1 TOWARD SOLUTION PROCESSED ORGANIC LIGHT EMITTING DIODES : ELECTRODE, TRANSPORT LAYERS, AND DEGRADATION STUDY By SHUYI LIU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2015
2 Â© 2015 Shuyi Liu
3 To my w ife
4 ACKNOWLEDGMENTS First, I thank my mentor, Dr. Franky So, for his trust, support and advise that guided me through 5 years of Ph.D study . I would like to thank my parents for their love and support. I also thank my wife , Xiaomeng, for her effort taking care of the family and her love which gives me power to overcome difficulties and challenges during the Ph. D study . A special thanks to all committee members , Dr. Franky So, Dr. Stephen Pearton, Dr. Jennifer Andrew, Dr. Rajiv Singh, and Dr. Andrew Rinzler, for their guidance, advice and passi on throughout the entire process of this graduate terminal project. I would like to thank all of the group members, who offered me inspiration, suggestion, and numerous help. Finally, I would like to thank all of the staff of the department for their help and technical support.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 6 LIST OF FIGURES ................................ ................................ ................................ .......... 7 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 14 1 INTRODUCTION ................................ ................................ ................................ .... 16 2 SOLUTION PRO CESSED SILVER NANOWIRE ELECTRODES .......................... 27 Preparation and Optimization ................................ ................................ ................. 27 Application in Solution Processed OLEDs ................................ .............................. 34 3 METAL OXIDE HOLE INJECTION/TRANSPORT LAYERS ................................ ... 44 Experiment Section for Nickel Oxide Film ................................ ............................... 45 Characterization of Nickel Oxide Film ................................ ................................ ..... 47 Hole Injection and Transport Properties of Nickel Oxide Film ................................ . 49 OLED Performance with Nickel Oxide HIL/HTL ................................ ...................... 52 Vanadium Oxide HIL ................................ ................................ ............................... 55 4 SURFACE PASSIVATION OF METAL OXIDE ................................ ....................... 73 Experiment Section ................................ ................................ ................................ . 75 PVP Passivation on NiO x HTLs and Its Application in OLEDs Devices .................. 76 AFM and XPS of the PVP Passivated NiO x Films ................................ ................... 79 PVP Passivation on VO x HILs and Its Application in OLEDs Devices .................... 84 PVP passivation on NiO x HTLs and Its Application in Solar Cell Devices ............... 85 5 DEGRADATION STUDY OF SOLUTON PROCESSED SMALL MOLECULE PHOSPHORESCENT OLED ................................ ................................ ................ 102 Experiment Section ................................ ................................ ............................... 103 Result a nd Discussion ................................ ................................ .......................... 104 LIST OF REFERENCES ................................ ................................ ............................. 123 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 131
6 LIST OF TABLES Table page 2 1 Performance of solution processed OLEDs with ITO and AgNWs electrodes. ... 37 3 1 Binding energies and component ratios of Ni 2p 3/2 and O 1s species ................ 58 3 2 Device characteristics of the solu tion processed OLEDs incorporating PEDOT:PSS HIL, as Prepared and UV O 3 treated NiO x (500 Â°C) HIL/HTL ....... 58 4 1 OLED performance incorporating as prepared NiO x , UVO NiO x , P NiO x , and P UVO NiO x HTLs. ................................ ................................ ............................. 88 4 2 The resolved XPS data of C1s, O1s and Ni 2p 3/2 signals of PVP passivated NiO x surface before and after UV O 3 treatment. ................................ ................. 89 4 3 The performance of MAPbI 3 perovskite solar cells incorporating PEDOT:PSS, as prepared NiO x , UVO NiO x , and P UVO NiO x HTLs . ................................ ...... 90 4 4 OPV performance incorporating PEDOT:PSS, as prepared NiO x , UVO NiO x , and P UVO NiO x HTLs. The standard deviation is reported in parenthesis . ....... 90
7 LIST OF FIGURES Figure page 1 1 The working principle of a typical OLED device. ................................ ................. 26 2 1 AgNWs coatings on glass substrates. ................................ ................................ 37 2 2 The fabrication process of AgNWs epoxy composite electrode using PDMS as the template. ................................ ................................ ................................ .. 37 2 3 The transparency measurement of AgNWs electrodes . ................................ ..... 38 2 4 Surface roughness analysis of AgNWs electrodes. ................................ ............ 39 2 5 Precisely patterned AgNWs electrodes after transferring to epoxy substrate. .... 40 2 6 Microscope image of the AgNWs electrodes pattern with a line width of approximately 100 m . ................................ ................................ ....................... 40 2 7 The schematic device structure and the molecular structures of all organic materials used in the solution processed OLED . ................................ ................ 41 2 8 Performance of solution processed OLEDs with ITO and Ag NWs electrodes. ... 42 2 9 Photographs of solution processed OLED s with ITO and AgNWs electrodes. ... 43 3 1 Transmittance (solid line) and absorbance (dashed line) spectra of a 40 nm thick solution processed NiO x film prepared on quartz substrates . ..................... 59 3 2 The surface morphology of NiO x under different treatment conditions. .............. 60 3 3 The XPS spectra of C 1s signals from 275 Â°C and 500 Â°C annealed NiO x . ........ 61 3 4 The XPS spectra of Ni 2p 3/2 and O 1s signals from 500 Â°C annealed NiO x before and after UV O 3 treatment . ................................ ................................ ...... 62 3 5 An ideal dark injection space charge limited current transient with circuit RC decay. ................................ ................................ ................................ ................. 62 3 6 Hole injection properties of NiO x with DI SCL measurement. ............................. 63 3 7 The field effect hole mobility measurement of NiO x . ................................ ........... 64 3 8 The I DS V DS output curves of the NiO x annealed at 275 Â°C . ................................ 65 3 9 The J V curves of the hole only devices with NiO x HTLs. ................................ ... 65
8 3 10 Performance of solution processed OLEDs with standard PEDOT:PSS or 500 Âº C annealed NiO x before and after UV O 3 treatment. ................................ .. 66 3 11 Performance of solution processed OLEDs with 275 Âº C annealed NiO x before and after UV O 3 treatment. ................................ ................................ ................. 67 3 12 Photographs of the EL of the PEDOT:PSS and NiO x devices taken in ambient atmosphere . ................................ ................................ .......................... 68 3 13 Performance of thermal evaporated OLEDs with PEDOT:PSS/NBP and 500 Âº C annealed NiO x after UV O 3 treatment. ................................ ........................... 69 3 14 Optical properties of solution processed VO x prepared on quartz substrates. ... 70 3 15 The AFM topography image of 40 nm solution processed VO x film prepared on ITO substrates. ................................ ................................ .............................. 70 3 16 The XPS spectra of V 2p 3/2 and O 1s signals of solution processed VO x . .......... 71 3 17 Hole injection properties of PEDOT:PSS and VO x . ................................ ............ 71 3 18 Solution processed OLEDs with VO x HIL. ................................ .......................... 72 3 19 Photographs of the EL of the PEDOT:PSS and VO x devices taken in ambient atmosphere . ................................ ................................ ................................ ........ 72 4 1 The PL of a 20 nm TCTA: 5wt% Ir(ppy) 3 EML deposited onto a 40 nm NiO x or TAPC HTL. ................................ ................................ ................................ ..... 91 4 2 OLEDs performance with TAPC or NiO x with different surface treatment. ......... 92 4 3 The EL spectra of a phosphorescent green OLEDs incorporating the as prepared NiO x HTL, UVO NiO x HTL, and P UVO NiO x HTL. ............................. 93 4 4 The PL of a 20 nm TCTA:5wt% Ir(ppy) 3 EML deposited onto 40 nm TAPC, as prepared NiO x , and UV O 3 treated NiO x . ................................ ....................... 93 4 5 The surface morphology of 40 nm as prepared NiO x , P NiO x , and P UVO NiO x . ................................ ................................ ................................ ................... 94 4 6 The AFM phase distribution histogram of 40 nm as prepared NiO x , P NiO x , and P UVO NiO x HTLs. ................................ ................................ ...................... 95 4 7 XPS spectra of the Ni 2p 3/2 signals from P NiO x and P UVO NiO x . .................... 96 4 8 XPS spectra of C 1s and O 1s signals from P NiO x and P UVO NiO x . ............... 97 4 9 OLED performance with as prepared and P UVO VO x HIL. ............................... 98
9 4 10 MAPbI 3 perovskite films formed on P UVO NiO x HTL. ................................ ....... 99 4 11 The performance of perovskite solar cells with PEDOT:PSS or NiO x w ith different surface treatment . ................................ ................................ ............... 100 4 12 The performance of OPVs with PEDOT:PSS or NiO x wit h different surface treatment. ................................ ................................ ................................ ......... 101 5 1 OLED architecture and mole cule structure. ................................ ...................... 113 5 2 Performance and operational stability of the OLEDs with solution processed or thermal evaporated EMLs. ................................ ................................ ........... 114 5 3 The accelerated degradation model. ................................ ................................ 115 5 4 Hole only devices with thermal evaporated or solution processed EMLs. ........ 116 5 5 PL d egradation of hole only devices with thermal evaporated or solution processed EMLS. ................................ ................................ ............................. 117 5 6 GC MS spectrum of both fresh toluene solvent and dated toluene solvent. The molecule structure of benzaldehyde is shown in the spectrum of dated toluene s olvent. ................................ ................................ ................................ 118 5 7 Voltage rise curves of hole only devices with or without UV excitation. ............ 119 5 8 Time resolved PL of the OLE Ds before and after degradation. ........................ 120 5 9 The PL EL intensity of the OLEDs with different EMLs after continuous driving for 25 min. ................................ ................................ ............................. 121 5 10 The two schematic degradation mechanisms for solution processed EMLs in hole only devices and OLED devices. ................................ .............................. 122
10 LIST OF ABBREVIATIONS 3TPYMB Tris[3 (3 pyridyl) mesityl]borane AFM Atomic force microscopy AgNWs Silver nanowires Al Aluminum electrode Alq3 Tris(8 hydroxyquinolinato)aluminum Au Gold BCP 2,9 dimethyl 4,7 diphenyl 1,10 phenanthroline BE Binding energy BHJ Bulk heterojunction Bphen 4,7 Diphenyl 1,10 phenanthroline C 60 Fullerene 60 CB Conduction band CBP 4,4' Bis(carbazol 9 yl)biphenyl CE Current efficiency DC Direct current DI SCL Dark injection space charge limited DMF Dimethyl formamide EBL Electron blocking layer E f Work function E g Band gap EIL Electron injection layer EL Electroluminescence EML Emitting layer EQE External quantum efficiency
11 E T Triplet energy ETL Electron transport layer FF Fill factor GC MS Gas chromatography mass spectrometry HBL Hole blocking layer HIL Hole injection layer HOMO Highest occupied molecule orbital HTL Hole transport layer IPA Isopropanol alcohol IQE Internal quantum efficiency Ir(mppy) 3 Tris[2 (p tolyl)pyridine]iridium(III) Ir(ppy) 3 fac tris(2 phenylpyridine)iridium (III) IS Impedance spectroscopy ITO Indium tin oxide electrode J sc Short circuit current J SCL Space charge limited current J t Transient current density J V L Current density voltage luminescence curve LiF Lithium fluoride LT 50 Time for the luminance to decay to 50 % of initial luminance LUMO Lowest unoccupied molecule orbital MAI Methyl ammonium iodide MAPbI 3 Iodide perovskite solar cell MoO x Molybdenum oxide N 2 Nitrogen
12 Ni(OH) 2 Nickel hydroxide NiOOH Nickel oxy hydroxide NiO x Nickel oxide NPB Di(1 naphthyl) diphenyl biphenyl) diamine OLED Organic light emitting diode OPV Organic photovoltaic PBD 2 (4 tert Butylphenyl) 5 (4 biphenylyl) 1,3,4 oxadiazole PbI 2 Lead iodide PC 60 BM (6,6) phenyl C61 butyric acid methyl ester PC 70 BM (6,6) phenyl C71 butyric acid methyl ester PCDTBT Poly(N 9Â´ heptadecanyl 2,7 carbazole alt 5,5 (4Â´,7Â´ di 2 thienyl 2Â´,1Â´,3Â´ benzothidiazole) PCE Power conversion efficiency PDMS Polydimethylsiloxane PE Power efficiency PEDOT:PSS poly(3,4 ethylenedioxythiophene) polystyrene sul fonate PET poly(ethylene terephthalate) PHJ Planar heterojunction PL Photoluminescence PMT Photomultiplier tube PVA polyvinyl alcohol PVK poly(N vinylcarbazole) PVP Polyvinylpyrrolidone RMS Root mean square roughness R s Sheet resistance
13 R se Serial resistance R sh Shunt resistance SAM Self assembly monolayer SEM Scanning electron microscope Si Silicon SiO 2 Silicon dioxide TAPC cyclohexylidenebis[N,N bis(4 methylphenyl)benzenamine] TCTA tris(4 carbozoyl 9 ylphenyl)amine TEG fac tris(2 (3 p xylyl)phenyl)pyridine iridium(III) TFT Thin film transistor T g Glass transition temperature TPBi 2,2',2" (1,3,5 Benzinetriyl) tris(1 phenyl 1 H benzimidazole) TPD N,N Bis(3 methylphenyl) N,N diphenylbenzidine TV Television UV O 3 Ultraviolet ozone VB Valence band V oc Open circuit voltage V on Turn on voltage VOTIP vanadium oxytriisopropoxide VO x Vanadium oxide XPS X ray photoemission spectroscopy XRD X ray diffraction
14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy TOWARD SOLUTION PROCESSED ORGANIC LIGHT EMITTING DIODES: ELECTRODE, TRANSPORT LAYERS, AND DEGRADATION STUDY By Shuyi Liu May 2015 Chair: Franky Fat Ke i So Major: Materials Science and Engineering Solution process ed organic light emitting diodes (OLEDs) have attracted much research attentions recently due to their great potential of lowering the fabrication cost. Although OLEDs with decent performance can be prepared with solution process able small molecule phosphorescent emitters, those devices are usually prepared on expensively sputtered electrodes , which is not favorable for lowering the fabrication cost. A nd the lack of appropriate solution process able hole transport layers (HTL) adds limitation to the fabrication of multi layer OLED devices with high efficiency. In terms of industrialization , the drawback in solution processed OLEDs is attributed to its substantiall y inferior operational stability as compared to t hermal evaporated counterparts. However, the underlying mechanism for the fast degradation of solution processed OLEDs is not yet well understood. This dissertation focus on the study of solution processed small molecule phosphorescent OLEDs in terms of three aspects : (I) preparation and application of novel low cost solution processed electrodes , (II) Preparation and application of novel solution processed functional layers , and (III) degradation study. In th e first section, fabrication of solution processed transparent silver nanowire s (AgNWs) electrode s and
15 their application in OLED devices are presented. With a novel patterning and transferring technique , the patterning issue, as well as the roughness and wetting problem s of AgNWs electrodes are solved . The OLEDs with AgNWs electrodes show high performance due to the scattering effect and elimination of shunt current . In the second section, the fabrication and application of solution processed metal oxides as an alternative hole inject ion and transport layers for solution processed OLED s are presented. Good hole injection and transport properties of the metal oxides are realized with optimized annealing conditions and surfa ce treatment. A novel and universal surface passivation method for metal oxides is presented to improve the performance of optoelectronic devices. In the last section, a comprehensive study is carried out to understand the fast degradation in solution proc essed emitting layers (EMLs) .
16 CHAPTER 1 INTRODUCTION AndrÃ© Bernanose and his co workers firstly discovered the electroluminescence ( EL ) in organic materials of acridine orange by applying high alternative voltages to the material, and they claimed this phenomena as the direct excitation of the dye molecules or of the electrons. 1 4 This discovery revealed a new and open era for making illumination devices with organic semiconductor materials . In 1960, o hmic injection electrodes to organic crystals of anthracene and observed its EL under DC bias using silver electrod es. 5 8 Five years later, W . Helfrich and W. G. Schneider reported EL observed in the anthracene crystal using electron and hole injecting counter electrodes , and this is the first time that the EL in organic materials is attributed to the recombination of injected electrons and holes . 9 In 1987, Ching W. Tang and Steven Van Slyke came up with the first prototype of organic light emitting diodes (OLEDs) , which wa s composed of two organic semiconductor layers that selectively transport electrons and holes . The electron hole recombination occu rred when the counter charge carriers m et each other in the middle of the organic layer. 10 The brilliant idea of separat ing electrons and holes injection and transport ha s significantly reduced the operating voltage and improved the e fficiency of the device , which paves the way for the emerging prosperous OLED research societies and industrialization in next few decades . A typical OLED ha s organic semiconductor layers deposited in between t wo counter electrodes. Based on the functionality of the organic semiconducting layers, the OLED s can have a hole injection layer (HIL), a hole transport layer (HTL), a hole blocking layer (HBL), an electron injection layer (EIL), an
17 electron transport lay er (ETL), an electron blocking layer (EBL), and an emitting layer (EML). A n electrode with deep work function (E f ) is used as the anode for the purpose of facilitating hole injection and a n counter electrode with shallow E f i s used as the cathode for the p urpose of facilitating electron injection. During electrical operation, holes are injected from anode into the valance band (VB) maximum of the organic layers and electrons are injected from cathode into the conduction band (CB) minimum of the organic layers. In organic semiconductors, t he VB maximum is called highest occupied molecular orbital (HOMO) and the CB minimum is called lowest unocc upied molecular orbital (LUMO) . The band gap (E g ) of the organic semiconductor is defined as the energy difference between its LUMO and HOMO level. When the holes and electron encounter each other within the EML, they tend to be weakly attracted to each other by electrostatic Coulomb forces. The bind states of electron hole pairs are called exc itons, and light emission is occurred when the excitons undergo radiative decay . The EL frequency/spectrum is corresponding to the relaxation energy of electrons in the excited state . D ue to the existence of exciton binding energy, the electron relaxation energy is slightly different from the band gap of emitter molecules. 11 A simple profile to demonstrate the working principles of OLED is shown in Figure 1 1 . T o characterize the performance of an OLED , several parameters are developed and defined as follows : The turn on voltage (V on ) rep resents the minimum voltage of applied bias required for EL emission; the current efficiency (CE) (unit: Cd A 1 ) and power efficiency (PE) (unit: lm W 1 ) are defined as luminous intensity per ampere in the normal direc tion and lum inous flux per watt. These are the two most commonly used parameters to represent OLED efficiency in commerce . To represent quantum efficiency
18 of the device, internal quantum efficiency (IQE) is defined as ratio of the number of total emitted p hotons per second to the number of charge carriers injected into the device per second, and external quantum efficiency (EQE) is defined as ratio of the number of photons escaped outside of the device per second to the number of charge carriers inje cted in to the device per second . As the most straightforward and prevalent parameter to characterize device efficiency, EQE is given by four independent factors: 12 r represents the fraction of excitons that quantum mechanically allowed to radiative decay, q eff (q) represents effective radiative decay ratio of the excitons that is quantum mechanically allowed to radiative decay , out represents the light out coupli ng efficiency , which is around 25% for most of the OLED devices without any light out coupling enhancement . 13 16 Therefore, to make a highly efficient OLED, each factor should be maximized to its theoretical limitation of one unity. Based on the quantum me chanics theory, a pair of spin 1/2 particles in the excited states can form either triplet states corresponding to a total spin angular momentum of S=1, or singlet states corresponding to a total spin angular momentum of S=0. The ratio of the number of tri plet excitons to the number of singlet ex citons is 3 to 1 based on probability and statistical calculation. In a fluorescent OLED, r is equal to 0.25 and the theoretically maximum EQE is 6% 7% without any light out coupling enhancement. The radiative decay is also very fast fluorescent OLEDs due to the nano second lifetime of singlet excitons. In a phosphorescent OLED, however, both singlet and triplet excitons can be harvest during light emission, r is equal to 1 and the theoretically
19 maximum EQE is 25% without any light out coupling enhancement. The radiative decay is also substantially slower in phosphorescent OLED due to the kinetically unfavorable occurred when the triplet states return to the ground singlet states. With good charge balance, efficient singlet and triplet excitons formation and harvesting , and appropriate light extraction techn ique, highly efficient small molecule phosphorescent OLEDs can be fabricated . For example, a maximum current efficiency exceeding 200 Cd A 1 has been reported in green phosphorescent OLEDs with micro cavity structures. 17,18 A record power efficiency of 120 lm W 1 has also been achieved in white OLEDs combining deterministic aperiodic nanostructures for broadband quasi omnidirectional light extraction and a multilayer energy cascade structure for energy efficient photon generation. 19 As the efficiency of sma ll molecule phosphorescent OLEDs has finally met the industrial demands to replace the prevalent inorganic light emitting display panel , they have prominent advantages in other aspects such as low er fabrication cost, reduced device thickness and weight, accordance with flexible panels and lower power consumption 20 23 . Recently, the generation and growth of OLED television (TV ) market further reveals the bright future of OLED industry . However, althou gh commercial available now, the prices of the OLED display panels are relatively high . For example, the price of the 55 inch OLED TVs is averaged more than $2000 in 2014. F urther technology innovation is in demand for companies to bring down the fabricat ion cost of OLED display panel, which turns out to be one of the biggest challenge s in OLED industries. As m ost of t he prevalent highly efficient OLEDs are fabricated using thermal evaporation method , which requires ultra high vacuum (< 1 Ã—
20 10 6 torr) level to prevent material degradation and contamination, this method lead s to high manufacturing cost with large material consumption . And it could be extremely difficult and expensive to make large area display panels due to the confinement of the vacuum chamber . Therefore, an alternative OLED fabrication process is required to significantly bring down the manufacturing cost and enable its application on large area devices . That is how people c a me to the idea of manufacturing OLEDs with solution process. To make an OLED device with solution process, the organic semiconductor materials with good solubility are first dissolved into appropriate solvent, the organic films are then deposited from solution precursors by spin coating, blade coating, spray printin g or inject printing. As the concentrations of these solution precursors are not high , the material consumption is usually very low in the deposition processes. And as the deposition processes are usually carried out in inert atmosphere, high vacuum chambers and power consumption thermal evaporation procedures can be spare d. The organic semiconductor films deposited using a solution process offers a great opportunity to bring down the OLED fabrication cost. However, there lies a hidden obstacle that hinders further reduction of fabrication cost in OLED industry most of the organic films are deposited onto the tin doped indium oxide (ITO) , which is the most commo nly used transparent anode in research and industry institutions . Although ITO electrode s are demonstrated with high conductivity (10 3 S cm 1 ) and transparency within the visible range (90% at 550 nm), 24,25 due to its expensiveness, ITO is not an ideal electro de for the purpose of cutting down the fabrication cost . One reason for the high price of ITO i s attributed to the scarcity of
21 indium and the skyrocketed price of indium in recent years; T he other reason is attributed to the exclusively us ed vacuum deposit ion method to fabricate ITO electrode , which adds up to the fabrication cost and limits its application to large area devices. 26 28 A nother drawback of ITO electrode is the incompatibility with flexible substrates due to its brittleness. To overcome these in herent problems of ITO electrodes, s ome solution process able conducting polymers such as p oly(3,4 ethylenedioxythioph ene) p olystyrene sulfonate (PEDOT:PSS) and its derivatives or modified counterparts have been developed and demonstrated as a potential transparent electrode can didate . But the performance of the optoelectronic devices with those conjugated polymer electrodes are usually limited due to their mediocre conductivity and transmittance . 29 3 5 Nano wires, on the other hand, offers a great opportunity for making low cost transparent electrodes as an alternative . A mong them the silver nanowire s (AgNWs) stand out due to the superior conductivity of silver. The AgNWs electrodes ha ve been demonstrated with comparable sheet resistance (R s and transparency (>80% withi n visible region) as the commercial ITO electrodes . 36,37 And t he AgNWs electrode prevails ITO in other aspect s such as large area application and abundance of silver resources in nature . Since the conductance of AgNWs electrode changes insignificantly with mechanical bending, it is also favorable for flexibility applications. However, as the AgNWs ha ve needle like feature s with diameters in nanometer scale and lengths in micrometer scale , the AgNWs directly deposited onto the substrates can easily penetrate into the thin organic films deposited on top and lead to severe shorting problem. Several strategies have been proposed to smooth the coarse surface of
22 AgNWs electrodes , and the most prominent results come from a polymer transferring process that significantly suppressed the surface roughness of AgNWs electrodes . This is achieved b y embedding the top surface of AgNWs on the substrates into a thick polymer matrix and transferring the AgNWs from the substrate to the polymer . As the inter face of the AgNWs in contact with the flat substrate is substantially smoother than its top surface, the surface roughness of the AgNWs electrodes is significantly reduced by exposing its original interface as the top surface for sequential film deposition . For example, X. Y. Zeng et. al . have reported transferring the AgNWs on a poly(ethylene terephthalate) ( PET) substrate to the surface of a polyvinyl alcohol (PVA) substrate. The roughness of the AgNWs drastically decreased from a root mean square ( RMS ) of 75 nm to a Rq of 1.5 nm after embedding it to the PVA surface. However, short circuit was still observed in thermal evaporated OLED s with AgNWs PVA electrode s even at small forward bias , and the current leakage could only be suppressed when a thick PE DOT:PSS layer is spin coated onto the AgNWs PVA to planarize its surface. 38 W. Gaynor et. al. demonstrated using mechanical pressure to transfer AgNWs on a glass substrate to the PEDOT:PSS layer. Due to the softness of PEDOT:PSS, the AgNWs will be embedded into PEDOT:PSS, and the roughness of the electrode is found decreasing with increasing PEDOT:PSS thickness. 39 Q. Pei et. al have fabricated efficient solution processed OLEDs with AgNWs polymer composite electrodes. The roughness problem of AgNWs electrod e is solved by embedding the AgNWs coated onto glass with a low viscosity polymer. During UV curing the polymer cross link together and form a rigid matrix. Since the polymer is hydrophobic with low surface energy , it is easily peeled off from the glass su bstrates. The OLEDs with AgNWs electrodes show higher efficiency
23 than the devices with ITO electrodes due to light scattering effect. 40,41 The bending test was performed on OLEDs with both AgNWs and ITO electrodes, with the result showing that AgNWs electr ode is more favorable for flexible purpose. However, th is process may have some intrinsic limitations : Since a hydrophobic polymer with low surface energy is required for successful peeling off from the glass substrates, the wetting on the AgNWs polymer substrates could be a problem for solution processed OLEDs. The electrode patterning/pixelization is also a common problem for solution processed AgNWs electrode, as most of the patterning procedures are manually controlled using a wet cotton tip or razor blade, the dimension precision and reproducibility of the electrode pattern is not reliable. Besides solution processed electrodes, a nother challenge to overcome in solution process OLED is realization of roll to roll fabrication process . Multilayer struc tures are usually required in small molecular phosphorescent OLEDs for good exciton confinement and thus high efficiency . W hile this is not a problem for thermal evaporated device s, it is a difficult to handle in solution processed films due to the dissolu tion problem. The paradox here is that good solubility of the materials is usually required for making solution processed OLEDs with reliable performance, however, good solubility also means the as deposited film is more prone to be damaged or dissolved by the solution used to deposit the following layer on top . One strategy to overcome the dissolution problem is to use orthogonal solvent system , which is achievable when the organic materials in two sequential layers have distinctive solubility in one type of solvent. This strategy is most commonly used to make solution processed EML/ETL bilayer structure. However, it should be noted that even with the orthogonal
24 solvent, the EML composed o f small molecule phosphorescent emitters and host can be mechanically damaged or removed by the orthogonal solvent spin cast on top . This is because most of the small molecules has low molecular weight and does not adhere to underlying layer strongly. Therefore, for most of the orthogonal solvent system a n large molecular w eight polymer host s uch as poly(N vinylcarbazole) (PVK) is used as matrix to prevent the mechanical wash away . 42 48 This leads to lower carrier mobility within the EML, and t his approach is also limited by the few choices of small molecule org anic materia ls that is soluble in polar solvent . For example, it is difficult to apply the orthogonal solvent strategy to make solution processed HTL/EML bilayer structures , where most of the materials used are only soluble in organic solvents. Therefore, the solution processed EMLs are usually directly deposited onto the PEDOT:PSS HIL. Due to the lack of electron blocking properties and strong exciton quenching nature of PEDOT:PSS, the solution processed OLEDs without any HTLs usually show lower efficiencies and a str ong efficiency roll off . To achieve the solution processed HTL/EML bilayer structure, o ne choice is to use cross linkable HTLs . A rigid and robust 3 D polymer network is formed usually via thermal annealing , 49 59 and some other cross linking procedure s such as radical/proton and photo initiated polymerization have also been reported. 60 63 Although the cross linkable HTL can with stand conventional organic and polar solvent attack, the hole mobility of the se resulting HTLs after cross linking is often re duced due to the more porous morphology compared to the thermally evaporated films, leading to lower efficiencies. On the other hand, as most of the solution processed OLEDs incorporate PEDOT:PSS as the HIL, the acidity and hygroscopic nature of PEDOT:PSS could lead to issues of device stability . 64,65 Therefore,
25 the development of alternative high mobility HIL and HTL materials compatible with solution processing is of utmost importance. The last challenge to overcome is the stability of solution processed small molecule phosphorescent OLEDs. As the s olution processed small molecule phosphorescent OLEDs have achieved comparable high efficiencies as their thermal evaporated counterparts, in terms of the stability, they have substantiall y shorter operational lifetimes . 66 Comprehensive studies on comparing the performance differences between solution processed and thermal evaporated EMLs are conducted, revealing that there are essential differences between the molecule packing density and orientation . 66,67 However, these differences are not sufficient enough to explain for the significantly faster degradation of the OLEDs with solution processed EMLs. The degradation in OLEDs can be divided into three categories: thermal induced degradation , electrical induced degradation and photo induced degradation. 68 In thermal induced degradation, the phosphorescent emitters tend to form aggregation in solution processed EMLs, and the phase segregation between the emitter and host during operation could cause the fast EL drop at constant current. 66 , 69 However, this degradation mechanism is less likely to occur when materials with high glass transition temperatures (T g ) and high solubility are used to prepare the EML. Therefore, a more detailed study on t he electrical and photo induced degradation and their effect on each other should be carried out to shed light on the culprit behind the scene for the fast degradation of solution processed small molec ule phosphorescent OLED s .
26 F igure 1 1. The working principle of a typical OLED device.
27 CHAPTER 2 SOLUTION PROCESSED SILVER NANOWIRE ELECTRODES In this work presented here , instead of spin casting the AgNWs suspension directly onto glass s ubstrates, where a hydrophobic polyme r matrix is required t o embed and transfer the AgNWs, the author presents a novel technique by spin coating AgNWs on to original hydrophobic template s , a hydrophilic polymer is thus can be used to transfer the AgNWs mesh, which significantly improve d the wetting properties of Ag NWs electrode . The challenges for s elective and precise patterning of AgNWs electrode can also be overcome with this technique. The AgNWs electrode made with this new approach has good wetting abilities, smooth surface roughness, precisely defined electrod e patterning, and comparable transmittance and sheet resistance as the commercial available ITO electrode. And due to the strong light scattering effect which enhanced the light out coupling efficiency, solution processed small molecule phosph orescent OLED s incorporating AgNWs electrodes show better performance with a 40% higher current efficiency compared to those incorporating commercial ITO electrodes. Preparation and Optimization AgNWs are directly purchased from Seashell Technology . The AgNWs used in this work have an average length of 10 (with a length distribution from 5 and an average diamete r of 60 nm (with a diameter distribution from 50 70 nm) . The AgNWs with small er feature sizes is intentionally chosen among all other pro ducts due to the empirically presumption that better scattering property can be achieved . The nanowires are originally dispersed in isopropanol alcohol ( IPA) solvent at a concentration of 20 mg mL 1 upon arrival, and the suspension was further diluted with
28 methanol. Figure 2 1 shows the scanning electron microscope (SEM) top image of the AgNWs directly spin casted onto pre cleaned glass substrates and the transmittance spectra of AgNWs with specified sheet resistance (R s ) . 70 As mentioned early, the AgNWs spin coated onto glass can have very coarse surface with protruding nanowires in a height scale of a few hundred nanometer s , which makes it inappropriate for OLED application. A transferring process is thus required to embed the rough surface of AgNWs into polymer matrix. This is usually done with AgNWs coated on glass substrates, where a hydrophobic polymer is used to embed the AgNWs and peel the electrode off from the glass after solidifying. The hydrophobicity of the polymer substrate would bri ng wetting problem for the solution processed devices with AgN Ws electrode , and precise patterning is also difficult to be realized using this method. Therefore, in this work, the author presents a novel transferring process: instead of transferring AgNWs directly from glass substrates, a recoverable hydrophobic buffer layer is deposited onto the glass substrates which serve as the template for AgNWs spin casting. Due to the low surface energy of the hydrophobic buffer layer, the surface adhesion between th e buffer layer s to any type of embedding polymer is weak and thus a hydrophilic polymer can be used as the matrix to embed the AgNWs . The precise pattern ing of the AgNWs electrode can be realized by selective modification of the surface energies of the hyd rophobic templates, and the selected areas serve as the temporarily wetting zones to confine the AgNWs spin coated onto the templates. One typical material that can be used as the recoverable hydrophobic buffer layer is the Polydimethylsiloxane (PDMS) resin . To prepare the PDMS buffer layer, glass substrates were first cleaned with detergent, deionized water, acetone and IPA for
29 10 min each using ultra sonication . The substrates were then put into an ultraviolet ozone (UV O 3 ) chamber for 20 min curing time. The viscous PDMS precursor was then taken out of a refrigerator and mixed with curing agent at a weight ratio of 9:1 for 5 min. After mixing the precursor was immediately cast onto the middle of the glass substrates for 3 4 drops using a sp atula , and the substrates were put onto a spin coater where a ramp speed of 1000 rpm for 1 s and a spin speed of 4000 rpm for 50 s takes place. A highly transparent and unifor m PDMS film is formed on the glass substrates after spin coating. The films were then annealed at 100 Âº C for 1 hour in ambient atmosphere for complete drying and solidification. After annealing, a rigid PDMS resin with highly hydrophobic surface is formed on the glass substrates. T he PDMS resin has been used as a stamp to transfer AgNWs from aluminum oxide membrane to flexible PET substrates by applying mechanical pressure . 7 1 ,7 2 However, the surface roughness of AgNWs electrodes is still significant on PET substrates and thus there is hardly any reports of optoelectronic devices with AgNWs electrodes prepared with PDMS stamps . In this approach, however, PDMS resin is used as the template for AgNWs deposition instead of the transferring st amp , and no mechanical pressure is required to transfer the AgNWs onto PDMS . Due to the flat PDMS surface in contact with the AgNWs mesh , the AgNWs electrode transferred to the polymer matrix should have significantly reduced surface r oughness. The difficu lty to overcome in this approach is to deposit uniform AgNWs mesh film on the PDMS template. As PDMS is highly hydrophobic with poor wetting properties , direct spin casting the AgNWs suspension onto the PDMS leads to the absence of AgNWs on PDMS surface. T herefore, PDMS was treated with UV O 3 to
30 break and functionalize its inert surface chemical bond for better wetting. The UV O 3 treatment was carried out using a Jelight Model 42 UV O 3 cleaner, and it turns out that the wetting of PDMS template is only significantly improved under UV O 3 exposure for more than 45 min. The improved wetting is attributed to the silica species formed on PDMS surface, which originates from the oxidized speci es of siloxane moieties after long time UV O 3 exposure. U niform film s thus can be deposited on PDMS template s with a spin coating method . However, it should be noted that this improved wetting properties of PDMS surface is only temporary, and either by agi ng in air for long time or by annealing the PDMS template at moderate temperatures would recover its surface energy back to its original value, and the treated PDMS surface becomes highly hydrophobic again. The recovery hydrophobicity nature of PDMS is att ributed to the migration of low molecule weight polymers from the bulk PDMS to top, covering the hydrophilic silica species on its surface . The recovered hydrophobicity of PDMS enables the use of hydrophilic polymers to transfer AgNWs for wetting purpose. To prepare the AgNWs electrode , the as received AgNWs suspension was diluted with methanol, and the diluted AgNWs suspension undergoes a short ultra sonication for 1 2 min to evenly disperse the AgNWs into the solvent before deposition. A longer ultra son ication may be detrimental to the AgNWs by breaking and rupturing the thin nanowires dispersed in solvent. Due to the recovery hydrophobicity nature of the PDMS template, the AgNWs suspension is immediately spin casted onto the PDMS templates after the UV O 3 treatment. The transmittance and sheet resistance of the AgNWs electrodes depend on the ir stac king densities on PDMS templates . For AgNWs with higher stacking density, smaller sheet resistance and lower transmittance is
31 observed, while for AgNWs with lower stacking density, larger sheet resistance and higher transmittance is observed. As the stacking density of AgNWs on top of PDMS is proportional to the concentration of AgNWs suspension, the volume of suspension dropped onto PDMS, and is inversely proportional to the volume dropping speed and spin speed, only AgNW s suspension concentration is varied while all of the other factors are fixed for good control of the AgNWs stacking density: the volume of AgNWs suspension spin casted onto PDMS is 2 0 micropipette to PDMS template is drop by drop in a time period of 5 s, and the spin speed of the substrate with PDMS template is fixed at 6000 rpm for 1 min . An annealing process is then ca rrie d out to drive away the residue solvent on AgNWs surface and to melt the cross section between two silver nanowires into a junction unit . T he junction formation significantly increase s the AgNWs electrode conductivity by 2 orders of magnitude by formin g a 2 D AgNWs network that allows in plane charge percolation while has little effect on its transmittance . Due to the lack of polyvinylpyrrolidone (PVP) binders for th e AgNWs used here, only moderate annealing temperature is required to form the AgNW junc tion, which is optimized to 140 ÂºC for 30 min. Further annealing would melt the AgNWs junctions to disconnected droplets due to localized heat, resulting in an increase of AgNWs sheet resistance. Due to thermal annealing, the surface of PDMS b ecomes hydrophobic again, and the AgNWs transferring process is done by covering the AgNWs mesh on PDMS template with hydrophilic UV adhesive polymers (in this case NOA 81 epoxy) and capping glass es . Due to the low viscosity of the liquid transferring poly mer, they penetrate into the void areas of the AgNWs mesh and form a flat interface with PDMS. The polymer s were then
32 solidified By UV curing and peeled off from the PDMS substrates. The schematic f abrication process of the AgNWs electrode is shown in Figu re 2 2. Figure 2 3 shows the transmittance spectra of the AgNWs electrodes and the photograph of the transparent AgNWs electrodes prepared from the precursor suspensions with three different concentrations corresponding to 2 mg mL 1 , 5 mg mL 1 , and 10 mg m L 1 . The commercial ITO with a standard sheet resistance ( R s = 20 + 5 ) is also compar ed in Figure 2 3. For the AgNWs electrode prepared from 2 mg mL 1 suspension, its sheet resistance (R s = 480 + is significantly larger than rest of the devices although it shows the highest transmittance among all of the AgNWs electrodes. The AgNWs electrode prepared from 10 mg mL 1 suspension shows even lower sheet resistance (R s = 12 + compared to the ITO electrode, however, the transmittance is low due to thick AgNWs stacking layer. Considering of the trade off between the transmittance and sheet resistance of the AgNWs electrodes, an optimized concentration of 5 mg mL 1 is used for the AgNWs precursor su spension. The corresponding AgNWs electrode has a comparable sheet resistance (R s = 30 + as the ITO electrode while still maintaining a high transmittance within the visible range . Another thing that should be noted is that all of the AgNWs electro des show similar transmittance from visible region to infrared, while the ITO shows much lower transmittance in infrared region. This result indicate s that AgNWs electrode s are more favorable for infrared applications than the commercial ITO electrodes due to their superior transmittance . The sheet resistance of the electrodes are measured with a four probe technique by depositing silver paint at the corners of the electrodes in a
33 square shape , the transmittance spectra of the electrodes are recorded using a PerkinElmer LAMBDA 950 UV/Vis/NIR Spectrometer. Figure 2 4 shows the atomic force microscopy (AFM) topography images of the AgNWs mesh (optimized from 5 mg mL 1 AgNWs suspension) prepared on a PDMS template before and after the epoxy transferring process . Tapping mode tip is used for the measurement. The . The section analysis of the film roughness is shown in Figure 2 4 as well. The AgNWs mesh prepared on PDMS before epoxy transferring process have very coarse surface with a pe ak to valley height of 4 82 nm and a R MS roughness of 75.2 nm due to the stacking and protrusion of nanowires. After the epoxy transferring process, the AgNWs electrode surface is drastically smoothed with a peak to valley height of 16 nm and a RMS roughnes s of 2.4 nm, which meet s the demands for the electrode application in optoelectronic devices. Except for the benefit from this approach that hydrophilic polymer matrix can be used to improve wetting of the AgNWs composite electrode, a precise/selective pat terning of the AgNWs electrode can also be realized by applying a shadow mask on top of PDMS during UV O 3 treatment. The UV O 3 treated area on PDMS then become hydrophilic while the other area remain hydrophobic, and during the AgNWs spin casing process, t he AgNWs mesh would only remain on those UV O 3 treated area that corresponds to the opening area of the shadow mask. Therefore, similar AgNWs electrode patterns as the openings of shadow mask is spontaneously formed on PDMS template after spin casting. Fig ure 2 5 shows the spontaneously formed AgNWs electrode pattern which has been transferred to the epoxy substrates. The electrode patterning shows good precision at the edge as revealed by the microscope image. A
34 micro scale patterning of the AgNWs electrod es is also achievable with this technique. Figure 2 6 shows the spontaneously formed micro patterned AgNWs electrode with a The well defined patterning, good wetting properties, as well as smooth surfaces of the AgNWs composite electr odes fabricated with this technique enable its further application to high performance solution processed small molecule phosphorescent OLEDs. Application in Solution Processed OLEDs The optimized AgNWs composite electrodes were used as the transparent anodes for solution processed small molecule phosphorescent OLEDs. For the device fabrication, a 40 nm PEDOT:PSS ( Clevios P VP Al 4083 ) layer was spin coated onto as prepared AgNWs composite electrodes spin speed was set at 4000 rpm for 50 s. The PEDOT:PSS film was then baked in air at 130 ÂºC for 30 min to remove the residue solvent. As the E f of AgNWs is 4.7 eV, the hole injection from AgNWs anode to the adjacent organic layers will encounter a significant barrie r height. Therefore, t he PEDOT:PSS bi functional layer serves two purposes here: 1. It serve as a HIL/anode buffer layer to tune the E f of AgNWs anode for better hole injection (the PEDOT:PSS has a deep er E f of 5.2 eV ); 2. It serve as an conducting medium to enable charge carrier transport between AgNWs junctions and hole injection from the void area of AgNWs network to adjacent organic layers . 73 The solution precursor for the EML is prepared by dissolving the small molecule host of 4,4' Bis(carbazol 9 yl)b iphenyl (CBP) and small molecule phosphorescent emitter of Tris[2 (p tolyl)pyridine]iridium(III) (Ir(mppy) 3 ) into toluene solvent. The weight ratio of CBP host to Ir(mppy) 3 emitter is 93 wt% : 7 wt% and the solution concentration is 10 mg mL 1 . After 2 h of continuous stirring with a magnitude stir bar , both small molecules are
35 completely dissolved in toluene and a clear light yellow solution is formed. The solution precursor wa s then spin coated onto the PEDOT:PSS HIL at a spin speed of 3000 rpm in a N 2 f illed glove box, resulting in a uniform EML with a thickness of 35 nm. The EML was then slightly baked at 60 ÂºC on a hot plate for 30 min to dry off the residue solvent. The baking temperature is kept low because the CBP host is easy to crystalize due to i ts low glass transition temperature (T g ). The samples were then transferred into an evaporator, where 50 nm 2,2',2" (1,3,5 Benzinetriyl) tris(1 phenyl 1 H benzimidazole) (TPBi) as the ETL/HBL, 1 nm lithium fluoride (LiF) as the EIL, and 100 nm aluminum (Al ) as the cathode were sequential deposited under a high vacuum level of 1 Ã— 10 1 torr. The device with exactly same architecture was also fabricated on commercial ITO anode as a control. The schematic device structure and the molecular structures of a l l organic materials used in th e solution processed OLED system are shown in Figure 2 7 . Both OLEDs with AgNWs and ITO anodes have large pixel area of 0.6 cm Ã— 1 cm, and t he devices were encapsulated in a N 2 filled glovebox with epoxy and capping glass to pre vent water/oxygen induced degradation during the measurement. For the OLED characterization, the current of the devices were measured with a Keithley 2400 source meter and the photocurrents were recorded using a Keithley 6485 picoammeter with a silicon ph otodiode. The EL spectra were measured with an Ocean Optics HR4000 spectrometer. Figure 2 8 A shows the current density voltage luminescence ( J V L ) curves of the solution processed OLEDs with AgNWs composite electrode (red) and ITO electrode (black) , Figur e 2 8 B shows the current efficiency curves of corresponding devices, and Figure 2 8 C shows the normalized EL spectra measured at a constant current density of 2.5 mA cm 2 . The performance data are
36 summarized in Table 2 1. Both devices with commercial ITO anode and AgNWs composite anode have same V on of 3.0 eV, indicating good hole injection properties of the electrodes with PEDOT:PSS HIL. Both devices also show very comparable current density curves due to the similar R s of AgNWs electrodes and ITO electro des. However, the solution processed OLED fabricated on AgNWs composite electrode shows an overall 4 0 % higher current efficiency than the device fabricated on commercial ITO electrode starting from a luminance intensity of 500 Cd m 2 . As the current effici ency enhancement of the solution processed OLED with AgNWs electrode remain the same, it indicat e that this enhancement should be correlated with enhanced light out coupling efficiencies. Figure 2 9 shows the light emission photos taken from both devices a t a constant current density of 2.5 mA cm 2 and photographs taken from the back side (glass substrate side) of both devices under incandescent light . O bvious light scattering phenomena was observed in OLED with AgNWs composite electrode . Due to the strong light scattering effect , substrate mode and waveguide mode in the OLED devices could be extracted, which explain ed for the higher efficiencies in AgNWs based OLEDs.
37 Table 2 1. Performance of solution processed OLEDs with ITO and AgNWs electrodes. Electrode V on Maximum CE CE at 1000 Cd/m 2 CE at 10,000 Cd/m 2 ITO 3.0 V 50 Cd A 1 36 Cd A 1 27 Cd A 1 AgNWs 3.0 V 55 Cd A 1 51 Cd A 1 41 Cd A 1 Figure 2 1 . AgNWs coatings on glass substrates. A ) SEM top view image. B ) Transmittance spectra of the AgNW s with specified sheet resistance. 70 Figure 2 2 . The fabrication process of AgNWs epoxy composite electrode using PDMS as the template. A B
38 Figure 2 3 . The transparency measurement of AgNWs electrodes. A ) Transmittance spectra of commercial ITO electrodes (black), AgNWs epoxy composite electrodes prepared from AgNWs suspension with concentration of 2 mg mL 1 (red), 5 mg mL 1 (blue), and 10 mg mL 1 (green) . B ) P hotograph of the transparent electrodes , June 11, 2011. Court esy of Shuyi Liu. From left to right: commercial ITO electrode, AgNWs electrode prepared from AgNWs suspension with concentration of 2 mg mL 1 , 5 mg mL 1 , and 10 mg mL 1 . A B
39 Figure 2 4 . Surface r oughness analysis of AgNWs electrodes. A) AFM topography image of the AgNWs deposited on PDMS . B) AFM topography image of the AgNWs transferred to epoxy substrate . C) The section roughness of AgNWs deposited on PDMS. D ) The section roughness of AgNWs transferred to epoxy substrate. A B C D
40 Figure 2 5. Precisely patterned AgNWs electrodes after t ransferring to epoxy substrate. A) Photograph of the AgNWs electrodes pattern, January 5, 2015. Courtesy of Shuyi Liu. B) Microscope image showing the sharp edge of the patterned AgNWs electrodes. Figure 2 6. Microscope image of the AgNWs electrodes pattern with a line width of approximately 100 m. A B
41 Figure 2 7 . The schematic device structure and the molecular structures of all organic materials used in the solution processed OLED .
42 Figure 2 8 . Performance of solution processed OLED s with ITO and AgNWs electrodes . A ) The J V L curves . B ) The current efficiency curves. C ) The EL spectra . A B C
43 Figure 2 9 . Photograph s of solution processed OLEDs with ITO and AgNWs electrode s , April 4, 2012. Courtesy of Shuyi Liu. A ) The light emission image with ITO electrode . B) The pixel image with ITO electrode. C) The light emission image with AgNWs electrode. D ) The pixel image with AgNWs electrode. The light emission photograp hs were taken under a constant current density of 2.5 mA cm 2 . The OLED fabricated on AgNWs electrode shows a strong scattering effect. A B C D
44 CHAPTER 3 METAL OXIDE HOLE INJECTION/TRANSPORT LAYERS In this chapter , instead of using organic materials, the author represents a new strategy of using solution processed metal oxides semiconductors as the functional layers in OLEDs. Compared to their organic counterparts, metal oxides have several advantages such as high carrier mobility , tunable energy level alignment and good air stability. Furthermore, a s most metal oxides are insoluble in organic solvent s , they are compatible with solution process ing for multilayer devices. Several metal oxides have already been shown to be solution processable , 74 80 making them pot ential candidates for roll to roll fabrication of OLED devices to reduce the fabrication cost. In this chapter, s olution processed nickel oxide (NiO x ) is first demonstrated as both a HIL and HTL in OLED devices . It is known that non stoichiometric NiO x is a wide band gap insulator with a room temperature conductivity of 10 13 S cm 1 . 81 On the other hand , non stoichiometric NiO x with a large density of Ni 2+ vacancies accompanied by compensation of holes or Ni 3+ , is one of the few p type metal oxide semi conductors with a good hole selectivity. 82 8 8 Recently, solution processed NiO x has been extensively studied as HTL s in organic photovoltaics (OPVs). 89 91 However , there are a very few report s on solution processed NiO x as a functional layer in OLEDs. Spec ifically, only sputtered NiO x has been attempted to be used as an HIL in tris(8 hydroxyquinolinato)aluminum (Alq 3 ) based fluorescent OLEDs. However, the se devices show ed very low efficiencies. 92 94 Here , the author demonstrate s high efficien cy solution processed phosphorescent OLEDs using solution processed NiO x as both an HIL and an HTL. By varying the annealing temperatures and treating the NiO x surface with UV O 3 , the author w as able to modify the morphology and chemical stoichiometry
45 of the NiO x surface , resulting in improve ment in hole injection and transport properties. The resulting solution processed phosphorescent green OLEDs with s olution processed NiO x show ed a maximum current efficiency of 70.0 + 1.6 Cd A 1 and a power efficiency of 75.5 + 1.8 lm W 1 , which was significantly higher than the device with a PEDOT:PSS HIL . The shelf stability of the OLEDs with NiO x HIL/HTL are also significantly improved compared to the devices with PEDOT:PSS. Solution processed vanadium oxide (VO x ) is then demonstrated as a HIL to replace the commonly used PEDOT:PSS in OLED devices. Like most solution processed metal oxide semiconductors, non stoichiometric VO x with a large density of O 2 vacancies accompanied by compensation of electrons or V 4+ , i s one of the n type metal oxide semiconductors. 95 97 Due to its deep CB energy level, the VO x HIL serve as an electron acceptor in OLED. Unlike solution processed NiO x which require high annealing temperature, solution processed amorphous VO x can be synth esized at room temperature, which is favorable for industrialization and application to flexible plastic substrates . T he solution processed OLEDs with VO x HILs show comparable performance as PEDOT:PSS devices. The shelf stability of the OLEDs with VO x HIL are also significantly improved compared to the devices with PEDOT:PSS . Experiment Section for Nickel Oxide Film The nickel oxide precursor was synthesized according to our previous publication. 6 5 To prepare the nickel ink precursor , nickel acetate tetrahydrate (Ni(CH 3 COO) 2 Â·4H 2 O) and ethanolamine (NH 2 CH 2 CH 2 OH) were dissolved in ethanol with a mol ar ratio of 1:1 . The film thickness was controlled by the concentration of the precursor ink . The solution was stirred in a sealed vial for 2 hours, and a h omogeneous and clear dark green solution was formed. A 40 nm thick NiO x film was formed via spin -
46 coating the precursor solution onto the pre cleaned substrates , followed by annealing at 275 Â°C for 1 hour. From the X ray Diffraction (XRD) data, t he resultin g films were polycrystalline . 65 To use the solution processed NiO x films for HILs/HTLs in OLEDs, the author further modified the annealing temperature to 500 Â°C for better crystallization of NiO x films. T he annealing process was performed in ambient air. After cooling down to room temperature, this solution processed NiO x film was loaded into a UV O 3 chamber for further surface treatment. The UV O 3 exposure time was around 5 10 min. There was no significant change of the film transmittance after UV O 3 treatment. To investigate the annealing temperature s and UV O 3 treatment effects on the surface properties of the solution processed NiO x films , AFM was used to examine the morphology, and X ray photoemission spectroscopy (XPS) measurements were carried out to study the surface chemistry. The hole injection efficiencies of s olution processed NiO x and PEDOT:PSS were characterized by dark injection space charge limited (DI SCL) transient measurement s . T he field effect mobility of solution processed NiO x was measured with a p channel thin film transistor (TFT). To fabricate solution processed OLEDs, th e solution processed NiO x films were deposited onto the ITO substrates. Th e substrates were then transferred into a nitrogen glove box. A mixture of CBP host and Ir( m ppy) 3 phosphorescent dopant in chlorobenzene solution was used for EML, with the weight ratio of 0.92 : 0.08. After spin coat ed onto PEDOT:PSS or solution processed NiO x , the EMLs were dried at 60 Â°C for 30 min . Afterwards, the samples were transferred into a n evaporator. T ris (2,4,6 triMethyl 3 (pyridine 3 yl)phenyl)borane (3TPYBM) , LiF and Al were vacuum deposited as ETL and cathode, respectively . To compare the performance of an OLED with an organic HTL, a 30 nm thick Di(1 -
47 naphthyl) diphenyl biphenyl) diamine (NPB) layer was deposited onto the PEDOT:PSS HIL by thermal evaporation. However, due to the difficulty of making a sol ution processed OLED with organic HTLs, the author thermal evaporated the emitting layer of CBP doped with 8 wt% of fac tris(2 phenylpyridine)iridium (III) (Ir(ppy) 3 ) onto both solution processed NiO x HIL/HTL and PEDOT:PSS HIL/NPB HTL . TPBi was deposited as a thin exciton /hole blocking layer and Alq 3 was deposited as the ETL. Characterization of Nickel Oxide Film T he nickel ink precursor was spin coated onto ITO substrates and annealed at either 275 Â°C or 500 Â°C . Figure 3 1 shows the transmittance and absorbance of the 500 Â°C annealed NiO x film on a quartz substrate. A high transmittance of above 90% over the entire visible region was obtained . From the absorption onset wavelength of 355 nm, the E g is determined to be 3.5 eV. The topography and phase images of the solution processed NiO x films are shown in Figure 3 2. The NiO x film annealed at 275 Â°C has an average crystalline grain size of 10 nm, with a RMS roughness of 0.87 nm. The NiO x film annealed at 500 Â°C sho ws larger crystalline grain size of 30 nm and a larger RMS roughness of 1.12 nm. The crystalline size s of the solution processed NiO x films under different annealing temperatures are in good agreement with th ose reported in the literature . 98 Interestingly, after a short UV O 3 treatment, the phase image of the 500 Â°C annealed NiO x film is found significantly changed with a lot of pinhole like patterns. This morpholog ical change corresponded to the increased film RMS roughness : from 1.12 nm to 2.66 nm. These results suggest that the UV O 3 treatment might lead to changes in surface composition of the solution processed NiO x film. To verify the surface chemistry of solution processed NiO x films , the surface compositions of the NiO x films is analyzed using XPS with the adventitious C 1s peak referenced to
48 284.8 eV . The C 1s signals of solution processed NiO x annealed at 275 Â°C and 500 Â°C are shown in Figure 3 3 . The main peaks in both C 1s signals are adventitious carbon peaks which are calibrated to a binding energy of 284.8 eV. The shoulder peak in C 1s signal of 275 Â°C annealed NiO x has a binding energy of 288.3 eV, which is corresponding to the acetate component in the nickel ink precursor. 99,100 This result revealed that the nickel ink precursor d id not fully decompose when annealed at 275 Â° C . Figure 3 4 shows the peaks of Ni 2p 3/2 orbital and O 1s orbital in solution processed NiO x films with and without UV O 3 treat ment. The measurement was performed at a take off angle of 20 Â°. The main peaks in the Ni 2p 3/2 and O 1s signals correspond to the stoichiometric NiO. To further investigate the species, t he shoulder peak in the O 1s signals was fitted with signals from nickel hydroxide (Ni(OH) 2 ), nickel oxy hydroxide (NiO(OH)) and w ater (H 2 O). Similarly, the shoulder peak in the Ni 2p 3/2 signals was fitted with three peaks , for Ni(OH) 2 , NiO(OH) and the nickel intersite respectively . 78,100 As shown in Table 3 1 , the UV O 3 treatment introduced more dipolar NiO(OH) species onto the film surface, which could result in a change in the surface dipole and hence a vacuum level shift at the inorganic organic interface , facilitat ing hole injection from NiO x to EML. 78 The optimum surface treatment time is 5 minutes and further treatments do not make significant change s . These findings are in good agreement with other reports on O 2 plasma treated NiO x films. 78,90 It should be noted that the H 2 O composition is doubled after the UV O 3 treatment. Indeed, another possibility is that the UV O 3 treatment improve s the wetting of NiO x and thus the adhesion to the adjacent EML , resulting in enhanced hole injection.
49 Hole Injection and Transport Properties of Nickel Oxide Film DI SCL transient measurements were conducted to compare the hole injection efficiency of solution processed NiO x HIL/HTL and the conventional PEDOT:PSS HIL. The hole only devices consist of HIL followed by a 2.0 m thick of NPB HTL. UV O 3 treated ITO and gold (Au) served as the counter electrodes. Figure 3 5 shows the long voltage step applied to a device and the transient current density ( J t ) characteristics in a DI SCL transient measurement. At time t = t DI, the current density reaches to its maximum value of J DI . For t > t DI , the current density begins to decay until it reaches to a steady state value of J SS . The ratio of J DI to J SS is expressed as : For an ideal o hmic injection with field independent condition, n equals zero and J DI /J SS equals 1.21, deviation from this ideal value of 1.21 indicates that the charge injection contact deviates from an Ohmic injection . To calculate the hole mobility and injection efficiency, the transient time t tr can be expressed as: The hole mobility is calculated from: Where V is the applied voltage and d is the film thickness. The space charge limited current density (J SCL ) can be calculated as: Where V is the applied voltage, d is the film thickness, Âµ 0 temperature dependent parameters derived from:
50 The hole injection efficiency can be calculated from: T he transient current density (J t) curves for devices with solution processed NiO x HIL annealed at 275 Â°C and 500 Â°C before and after UV O 3 treatment, along with the conventional PEDOT:PSS (P VP Al4083 from Clevios TM ) HIL are presented in Figure s 3 6 A D , respectively . The ratio of J DI to J ss is kept around 1.21 for both PEDOT:PSS and UV O 3 treated NiO x samples under different applied voltages, indicating a n Ohmic contact for holes. 101,102 The NiO x annealed at 500 Â°C followed by the UV O 3 treatment shows the best hole injection with the highest J ss . Figure 3 6 E shows the hole mobility of NPB and Figure 3 6 F shows the injection efficiency of NiO x annealed at 500 Â°C followed by the UV O 3 treatment and PEDOT:PSS . The injection efficiency of the UV O 3 treated NiO x is about 0.8, which is significantly higher than that of PEDOT:PSS. However, without UV O 3 trea tment, the solution processed NiO x film shows a lower hole injection efficiency . The 275 Â°C annealed NiO x has the lowest hole injection among all samples, which could be attributed to the low work function of the incompletely decomposed nickel acetate prec ursor components . And since the lower temperature annealed NiO x has a stronger chemical adsorption property, 103 the carbonaceous and hydroxyl species adsorbed on its surface could also result in a reduction of work function and thus hole injection efficiency. 104 106 In addition to the hole injection property, the hole transport property was studied. T he field effect mobility of the solution processed NiO x film was measured using a p -
51 channel TFT. A 100 nm thick solution processed NiO x film is spin co ated onto the silicon/silicon dioxide ( Si/SiO 2 ) substrates with a 300 nm thick SiO 2 (capacitance: 10 nF cm 2 ) and annealed at either 275 Â°C or 500 Â°C , followed by a 60 nm thick Au evaporated as the drain and source contacts. The TFT has a channel length of 1 mm. Figure 3 7 shows the device structure and the TFT output current data . The field effect mobility is calculated from: Where I DS is the source drain saturation current, V G is the applied gate voltage, W is the channel width, and Ci is the capacitance of the SiO 2 . For the solution processed NiO x annealed at 500 Â°C, the slope extracted from the I DS 1/2 V G curve is 1.33 Ã— 10 4 , corresponding to a field effect mobility of 0.141 cm 2 V 1 s 1 , which is significantly higher than the value of organic HTLs. This mobility is also comparable to the reported values of 0.43 cm 2 V 1 s 1 for the sputtered nickel metal films followed by a similar oxidization pr ocess (500 Â°C annealing in air ) . 107 However, for the NiO x film annealed at 275 Â°C, the TFT source drain current level was low and no saturation region was found as shown in Figure 3 8 . This is possibly due to the smaller crystalline grains of the solution processed NiO x annealed at low t emperatures . The grain boundar ies in polycrystalline films scatter carriers and hinder the carrier transport. Therefore, it is expected that 500 Â°C anneale d NiO x exhibits a higher hole injection efficiency and hole mobility than th at for the film ann e aled at 275 Â°C . It should also be noted that the bulk transport properties of the solution processed NiO x films are not affected by the UV O 3 treatment, which is proved by the similar J V curves of the NiO x hole only devices as shown if Figure 3 9 . To prepare th e hole only device, UV O 3 treated ITO and Au were
52 used as two deep work function electrodes. A 100 nm thick s olution processed NiO x film was formed by spin coating the nickel ink precursor (0.4 M) at 1000 rpm, followed by annealing at either 275 Â°C or 500 Â°C for 1 h. The 500 Â°C annealed NiO x shows higher hole current density than the 275 Â°C annealed NiO x . However, after UV O 3 treatment, no significant change of the hole current densities was found in the hole only devices . This is withi n our expectation sin ce the UV O 3 treatment only reaches to a few nanometers within the NiO x top surface, while the transport performance is related to the film bulk properties. T herefore , it is conclude d that the UV O 3 treatment plays a dominant role of altering the s olution processed NiO x film surface species, which significantly improves its hole injection efficiencies while has little effect on its hole transport properties . OLED Performance with Nickel Oxide HIL/HTL Encouraged by the enhanced hole injection and transport in the 500 Â°C annealed NiO x film , OLEDs with s olution processed NiO x HIL/HTL were fabricated. Figure 3 10 A shows the device architecture and the band diagram of the solution processed phosphorescent green OLEDs fabricated with PEDOT:PSS HIL or 500 Â°C annealed NiO x HIL/HTL before and after UV O 3 treatment. 18,108 Figure 3 10 B shows the device J V L curves and Figure 3 10 C shows the device CE and PE curves. To check the reproducibility of the device performance, 12 devices of each group were fabricated and measured, with the statistics of device performance summarized in Table 3 2. The EL spectr a of the devices are shown in Figure 3 10 D . With the as prepared NiO x HIL/HTL, the OLED devices show a larger V on compared to the devices with PEDOT:PSS due to the inefficient hole injection. The devices also show a relatively low power and current efficiency. After UV O 3 treatment on the solution processed NiO x HIL/HTL, the device
53 performance is significantly improved due to the enhanced hole injection. The devices have the lowest V on with a maximum power efficiency of 75.5 + 1.8 lm W 1 and current efficiency of 70.0 + 1.6 Cd A 1 , which are significantly higher than those of the devices with PEDOT:PSS. OLED devices were also fabricated with solution processed NiO x annealed at 275 Â°C . The resulting devices have a higher V on and a low er power efficiency. As the low temperature proce ssed NiO x surface is rich of Ni(OH) 2 species, 65 which is easily turned into NiO(OH) upon oxidation , the 275 Â°C annealed NiO x film turned much darker after UV O 3 treatment. The OLED devices with the 275 Â°C annealed NiO x films followed by UV O 3 treatment sho w reduced turn on voltages due to the improved hole injection. However, the devices still show an overall poorer performance compared to the devices with 500 Â°C annealed NiO x HTLs . Figure 3 11 shows the device performance with the solution processed NiO x HIL/HTL annealed at a lower temperature of 275 Â°C, before and after UV O 3 treatment. The device shows a larger turn on voltage of 4.5 V before UV O 3 treatment. Both devices show reduced current efficiency and power efficiency compared to the NiO x annealed at 500 Â°C . This could be attributed to larger amount of defect states within the low temperature processed NiO x films, and the interfacial gap states could also serve as the quenching sites for the excitons formed in the EML. Therefore, the high annealing temperature and UV O 3 treatment are obviously the key to improve the device performance. To evaluate the device shelve life , solution processed OLEDs with a PEDOT:PSS HIL and UV O 3 treated NiO x HIL/HTL were both driven at a constant current of 1 mA/cm 2 . W ith both devices encapsulated ; the devices were electrical ly driv en in ambient condition . Photos of the active areas of these devices are presented
54 in Figure 3 1 2 . Dark spots appear ed in the PEDOT: PSS device within two weeks and grew with time. The short shelf life of the PEDOT:PSS device is due to the moisture uptake during storage, resulting in decreased hole injection and luminance degradation. 65 On the other hand, the NiO x devices showed no dark spots after 6 weeks of storage in ambient . Thus , it is a clear demonstration that solution processed NiO x device outperforms conventional PEDOT:PSS one in terms of environmental stability . Since PEDOT:PSS only serve s as a HIL in solution processed OLEDs while NiO x serve s as both HIL and HTL, a further compar ison of solution processed NiO x HIL/HTL with conventional evaporated organic HTLs provides a more complete study . Here, the author conducted a comparison between solution processed NiO x and thermal evaporated HTL . The devices studied have the following str uctures: ITO/30 nm PEDOT:PSS/30 nm NPB /40 nm CBP: 8 wt% Ir(ppy) 3 /10 nm TPBi /30 nm Alq 3 /1 nm LiF/100 nm Al and ITO/40 nm UV O 3 treated NiO x /40 nm CBP: 8 wt% Ir(ppy) 3 /10 nm TPBi /30 nm Alq 3 /1 nm LiF/100 nm Al. The device p erformances are shown in Figure 3 1 3 A B . The EL spectra are shown in Figure 3 1 3 C . Compared to the reference device with a PEDOT:PSS HIL/NPB HTL, the device with solution processed NiO x HIL/HTL shows a similar J V L curve, with a PE higher than the reference device at low EL intensities and a comparable efficiency as the reference device at high EL intensities. These results show that the performance of UV O 3 treat ed NiO x as HIL/HTL for OLEDs is comparable to conventional thermal evaporated HTLs . The study of solution processed NiO x HIL/HTL demonstrate a pathway toward highly efficient solution processed multi layer phosphorescent OLEDs.
55 Vanadium Oxide HIL To prepare the V O x films, vanadium oxytriisopropoxide (VOTIP) is mixed with IPA solvent at different volume ratios, and the precursor solution is then spin coated onto pre cleaned ITO substrates. A uniform and highly transparent film with slightly yellow color is formed o n the ITO substrates, and no post annealing is required to solidify the film since the IPA solvent is volatile and evaporate fast during the deposition. The chemical reaction to form VO x is fast in air due to the existence of moisture. The chemical reactio n formula can be written as: OV[OCH(CH 3 ) 2 ] 3 + 3/2 H 2 O 1/2 V 2 O 5 + 3 (CH 3 ) 2 CHOH Figure 3 14 shows the transmittance and absorption spectra of a 40 nm VO x deposited on a quartz substrate. The VO x HIL is highly transparent with a direction forbidden transition bandgap of 2.25 eV. AFM was carried out to study the surface roughness of as deposited VO x films. The solution processed VO x film has a very uniform surface with a RMS roughness of 1 . 10 nm as shown in Figure 3 15, and the film is highly amorphous as confirmed by XRD. S urface compositions of the solution processed VO x films is then analyzed using XPS with the adventitious C 1s peak referenced to 284.8 eV . Figure 3 1 6 shows the peaks of V 2p 3/2 orbital and O 1s orbital in solution processed VO x films . The measurement was performed at a take off angle of 20 Â°. I n the V 2p 3/2 signals , the main peak with a binding energy of 517.5 eV correspond s to the stoichiometric V 2 O 5 , and the shoulder pe ak with a binding energy of 516.2 eV correspond s to the V 4+ species. The molecule ratio of V 5+ to V 4+ is 4:1 calculated from the measurement. The XPS result confirmed that the dominant defects in solution processed VO x HIL should be oxygen vacancy or vanadium interstitial, which is compensated by electrons or V 4+ . 95 97
56 DI SCL transient measurements were conducted to compare the hole injection efficiency of solution processed VO x HIL and the conventional PEDOT:PSS HIL. T he hole only devices consist of HIL followed by a 2.0 m thick of NPB HTL. UV O 3 treated ITO and Au served as the counter electrodes. Figure 3 1 7 shows the calculated hole mobility of NPB on VO x (red) or PEDOT:PSS (black) HIL and the hole injection efficie ncy from VO x (red) or PEDOT:PSS HIL to NPB. The room temperature processed VO x HIL shows a very similar injection efficiency as PEDOT:PSS. Solution processed OLEDs were then fabricated on top of VO x or PEDOT:PSS HILs. A mixture of CBP as the host, Ir( m ppy) 3 as the phosphorescent emitter , N,N Bis(3 methylphenyl) N,N diphenylbenzidine (TPD) as the hole transport material, 2 (4 tert Butylphenyl) 5 (4 biphenylyl) 1,3,4 oxadiazole (PBD) as the electron transport material were dissolved in chlorobenzene solution at a weight ratio of 0. 60 : 0.0 6 : 0.10 : 0.24 to prepare the solution processed EML . After spin coat ed onto PEDOT:PSS or VO x , the EMLs were dried at 60 Â°C in a N 2 filled glove box for 30 min . Afterwards, the samples were transferred into a n evaporator. 4,7 Diphenyl 1,10 phenanthroline (Bphen) , LiF and Al were vacuum deposited as ETL/HBL and cathode, respectively . Figure 3 1 8 shows the device structure with energy levels and the current efficiency curves. 108 Both devices fabricated on the PEDO T:PSS and VO x shows comparable current efficiency due to their similar hole injection properties. To evaluate the device shelve life , solution processed OLEDs with a PEDOT:PSS HIL and VO x HIL were both driven at a constant current density of 1 mA/cm 2 . With both devices encapsulated ; the devices were electrical ly driv en in ambient condition . Photos of the active areas of these devices are presented in Figure 3 1 9 . Dark spots appear ed in the PEDOT: PSS device within two
57 weeks and grew with time. On the other hand, just like NiO x , the VO x devices showed no dark spots after 6 weeks of storage in ambient . Thus , it is a clear demonstration that device with solution processed VO x also outperforms conventional PEDOT:PSS in terms of environmental stability .
58 Table 3 1. B inding e nerg ies and c omponent r atios of Ni 2p 3/2 and O 1s s pecies Related Components Ni 2p 3/2 Binding Energy (eV) O 1s Binding Energy (eV) R atio (from Ni 2p 3/2 ) R atio (from O 1s) Stoichiometric Nickel Oxide [ NiO] 854.1 a // 854.1 b 529.5 // 529.5 63.3% // 58.0% 66.0% // 54.6% Nickel Hydroxide [Ni(OH) 2 ] 855.6 //855.5 531.0 // 530.9 21.4%// 17.9% 16.9% // 14.9% Nickel Oxy hydoxide [NiO(OH)] 856.1 // 856.1 531.9 // 531.8 15.3% // 24.2% 14.7% // 23.1% Absorbents [H 2 O, O 2 ] 533.1 // 533.0 2.4% // 7.4% a The numbers before double slash ( // ) is acquired from the as prepared NiO x , b The numbers after double slash is acquired from the UV O 3 treated NiO x . Table 3 2. Device c haracteristics of the s olution p rocessed OLEDs i ncorporating PEDOT:PSS HIL, a s Prepared and UV O 3 t reated NiO x (500 Â°C) HIL/HTL Devices V on (V) V at 10 3 Cd m 2 (V) Maximum CE (Cd A 1 ) CE at 10 3 Cd m 2 (Cd A 1 ) Maximum PE (lm W 1 ) PE at 10 3 Cd m 2 (lm W 1 ) PEDOT:PSS 3.23 + 0.05 6.04 + 0.03 61.2 + 1.3 47.1 + 1.4 43.2 + 1.0 24.5 + 0.7 As prepared NiO x 3.80 + 0.02 7.40 + 0.07 55.3 + 2.3 24.7 + 1.4 42.1 + 2.2 10.5 + 0.7 UV O 3 treated NiO x 2.55 + 0.03 5.94 + 0.04 70.0 + 1.6 40.3 + 1.7 75.5 + 1.8 21.2 + 1.0
59 Figure 3 1 . Transmittance (solid line) and absorbance (dashed line) spectra of a 40 nm thick solution processed NiO x film prepared on quartz substrates .
60 Figure 3 2. The surface morphology of NiO x under different treatment conditions . A ) AFM phase image of 275 Â°C annealed NiO x . B ) AFM phase image of 500 Â°C annealed NiO x . C) AFM topography image of 500 Â°C annealed NiO x before UV O 3 treatment . D) AFM topography image of 500 Â°C annealed NiO x after UV O 3 treatment . E) AFM phase image of 500 Â°C annealed NiO x before UV O 3 treatment. F) AFM phase image of 500 Â°C annealed NiO x after UV O 3 treatment. A B C D E F
61 Figure 3 3. The XPS spectra of C 1s signals from 275 Â°C and 500 Â°C annealed NiO x . A) The C 1s signal from 275 Â°C annealed NiO x . B) The C 1s signal from 500 Â°C annealed NiO x . The take off angel is 20Â° B A
62 Figure 3 4. The XPS spectra of Ni 2p 3/2 and O 1s signals from 500 Â°C annealed NiO x before and after UV O 3 treatment . A ) The Ni 2p 3/2 signal from NiO x before UV O 3 treatment. B ) The O 1s signal from NiO x before UV O 3 treatment. C ) The Ni 2p 3/2 signal from NiO x after UV O 3 treatment. D ) The O 1s signal from NiO x after UV O 3 treatment. Figure 3 5 . An ideal dark injection space charge limited current transient with circuit RC decay. D A B C
63 Figure 3 6. Hole injection properties of NiO x with DI SCL measurement . A ) The transient current density with 275 Â°C annealed NiO x . B ) The transient current density with 500 Â°C annealed NiO x . C ) The transient current density with 500 Â°C annealed NiO x followed by UV O 3 treatment . D ) The transient current density with PEDOT:PSS . E ) The NPB hole mobility extracted from the devices with PEDOT:PSS (black square) and UV O 3 treated NiO x (red circle) . F ) The hole injection efficiency curves of PEDOT:PSS (black square) and UV O 3 treated NiO x (red circle) . A B C D E F
64 Figure 3 7. The field effect hole mobility measurement of NiO x . A ) The structure of NiO x thin film transistor . B ) the I DS V DS output curves . C ) the I DS(saturated) 1/2 V G transfer curve at V DS = 40 V. The linear fit of the curve yields a slope value of 1.33 Ã— 10 4 , corresponding to a hole mobility of 0.141 cm 2 V 1 s 1 . A B C
65 Figure 3 8 . The I DS V DS output curves of the NiO x annealed at 275 Â°C . Figure 3 9 . The J V curves of the hole only devices with NiO x HTLs.
66 Figure 3 10. Pe rformance of solution processed OLEDs with standard PEDOT:PSS or 500 Âº C annealed NiO x before and after UV O 3 treatment . A ) Device structure and energy levels with respect to vacuum level of the materials used in solution processed OLEDs. B ) J V L characteristics. C ) current eff iciency/power efficiency curves. D ) The EL spectrum. D A B C
67 Figure 3 11. Performance of s olution processed OLED s with 275 Âº C annealed NiO x before and after UV O 3 treatment . A ) J V L curves. B ) current efficiency /power efficiency curve s . A B
68 Fig ure 3 1 2 . Photographs of the EL of the PEDOT:PSS and NiO x devices taken in ambient atmosphere , September 16, 2013. Courtesy of Shuyi Liu.
69 Fig ure 3 1 3 . Performance of t hermal evaporated OLED s with PEDOT:PSS/NBP and 500 Âº C annealed NiO x after UV O 3 treatment . A ) J V L characteris tics . B ) current efficiency / power efficiency curves . C ) EL spectra . A B C
70 Fig ure 3 14. Optical properties of solution processed VO x prepared on quartz substrates. A ) A bsorption spectr um of a 40 nm VO x film . B) Transmittance spectrum of a 40 nm VO x film. Fig ure 3 15. The AFM topography image of 40 nm solution processed VO x film prepared on ITO substrates . The amorphous VO x film is highly smooth with a RMS roughness of 1.10 nm. A B
71 Fig ure 3 16. The XPS spectra of V 2p 3/2 and O 1s signals of solution processed VO x . A ) V 2p 3/2 XPS acquisition for V O x without further annealing. B ) O 1s XPS acquisition for V O x without further annealing. The take off angle is 20 Âº Fig ure 3 1 7 . Hole injection properties of PEDOT:PSS and VO x . A ) The NPB hole mobility extracted from the devices with PEDOT:PSS (black square) and solution processed V O x (red circle) . B ) The hole injection efficiency curves of PEDOT:PSS (black square) and solution processed V O x (red circle) . A B A B
72 Fig ure 3 18. Solution processed OLEDs with VO x HIL. A) The energy levels of the device. B ) the current efficiency of the devices with solution processed VO x HIL (red circle) and PEDOT:PSS (black square). Fig ure 3 19. Photographs of the EL of the PEDOT:PSS and V O x devices taken in ambient atmosphere , October 5, 2013. Courtesy of Shuyi Liu. A B
73 CHAPTER 4 SURFACE PASSIVATION OF METAL OXIDE Although Solution processed metal oxides functional layers offer a great opportunity for making highly efficient multi layer solution processed OLEDs , they have some intrinsic problems to overcome: As most of the solution processed metal oxides are synthesized in air, their surfaces are known to be rich of hydroxyl speci es which act as exciton quenching sites affecting the device performance. 109 111 In Chapter 3 , while efficient solution processed OLEDs incorporating solution processed NiO x can be demonstrated, the efficiency roll off is very strong when NiO x is used as an HTL. Contradictory to other reports that the efficiency roll off is due to the poor charge balance, 86,93,94 the author found that it is actually due to strong quenching at the NiO x HIL/HTL interface . T o alleviate the quenching problem, strat egies such as changing the carrier profile in the active layers by modifying the injection layers 110 and inserting an exciton blocking layer to spatially separate the active layer and the metal oxide layer have been used . 109,111 However, i nstead of address ing the quenching problem directly, these strategies were to keep the exciton forming zone away from the metal oxide surface, which complicates the device architecture and adds limitation to its application. Therefore, a more favorable and straightforward way to suppress exciton quenching is to passivate the surface of the metal oxide s . One strategy is to use self assembly monolayers (SAMs) on top of the oxides. 112 116 However, most SAMs are not designed for device applications, and they are especially unfavorable for solution processed optoelectronic devices due to the hydrophobic surface of the SAM layer that is problematic for the wetting process. 114 116 Therefore, t o passivate the metal oxide surface for solution process, a hydrophilic polymer is required and PVP is a potential
74 candidate. 117 120 The advantages of PVP polymer are : (i) good complexion ability wit h transitional metal ions , 119 , 121 , 122 (ii) good solubilit y in both organic and polar solvent s due to its amphipathic propert ies ; and (iii) large band gap energy (E g ~ 5.6 eV 119 ) . However, the challenge of using PVP as a passivation layer is that it is an insulating polymer prohibiting carrier transport and injec tion/extraction, which is inappropriate for optoelectronic device applications. As a result, there is hardly any report of using PVP as a passivation layer to make efficient optoelectronics. In this chapter , the author demonstrate s a novel strategy of usin g PVP as passivation layer on top of metal oxides to suppress ex cit on quenching and make highly efficient OLED and solar cell devices . While PVP itself is an insulating polymer, b y further treating the PVP passivation layer with UV O 3 , carrier injection from the PVP passivation layer can be drastically improved , resulting in substantially enhanced device performance . Here, the author chose NiO x (as a n HTL) and VO x (as a n HIL) to study the PVP passivation effect, and found that PVP is effective to passivat e both metal oxide surfaces to suppress the exciton quenching . To facilitate charge injection, the PVP passivated metal oxides are treated with UV O 3 , resulting in the formation of a strong chemical binding between PVP and the metal oxides surface species which enables charge injection. As a result, phosphorescence green OLEDs incorporating the PVP passivated NiO x HTLs followed by subsequent UV O 3 treatment show a high current efficiency of 90.8 + 2.1 Cd A 1 , which, to the best of our knowledge, is the highest in all OLEDs incorporating solution processed metal oxides as a carrier transport layer. Furthermore, OLEDs using NiO x as an HTL or VO x as an HIL show significantly reduced efficiency roll off with a PVP passivation layer. In addition to OLEDs, thi s approach can
75 also be used to improve the performance of perovskite solar cells incorporating solution processed NiO x HTLs, yielding a significantly enhanced power conversion efficiency (PCE) of 10.9 + 0.3%. Since this passivation technique is fully compati ble with solution processing, it implies that this approach is ubiquitous for metal oxides used in solution processed optoelectronic devices. Experiment Section The NiO x and VO x films were synthesized using the same recipe as reported in Chapter 3. The met al oxide thin films were prepared by spin coating the precursor solution onto the appropriate substrates. For XPS measurements , substrates were single crystal Si wafers. For all other measurements, substrates were pre cleaned ITO . After spin coating, the sample s were heated to 500 Â°C for NiO x and 150 Â°C for VO x . The as prepared NiO x film is a p type semiconductor with a polycrystalline structure confirmed by XRD , 65 and the as prepared VO x film is an n type amorphous semiconductor. 95 97 To passivate the met al oxide surface, PVP was dissolved into c hloroform and spin casted onto the metal oxide samples. The samples were then spin rinsed with chloroform, yielding an ultra thin (< 5nm) PVP passivation polymer. A short UV O 3 treatment (5 10 min) was subsequently carried out on the PVP passivated NiO x or VO x surface. There was no significant change of the film transmittance after PVP passivation and UV O 3 treatment. To investigate the exciton quenching properties of NiO x HTL and VO x HIL, the photoluminescence (PL) spectra of EMLs composed of a 20 nm thick tris(4 carbozoyl 9 ylphenyl)amine (TCTA) doped with 5wt% Ir(ppy)3 deposited on different samples were recorded with a monochromatic excitation wavelength of 350 nm. The role of PVP passivation polymer on suppressing the exciton quenching of metal oxides was
76 revealed by comparing the PL intensities of the deposited EMLs to that of the reference sample, which is prepared by depositing the same EML on top of a cycl ohexylidenebis[N,N bis(4 methylphenyl)benzenamine] ( TAPC) thin film . Due to its large Eg and high triplet energy (E T ), TAPC can effectively block the singlet and triplet excitons, forming a non quenching interface with the phosphorescent emitter. 123 125 To investigate the exciton quenching effects in OLED performance , p hosphorescent green OLEDs were fabricated on different samples. A dual EML was used for easy tuning of the charge balance and confining of the emitting zone, and was composed of a 20 nm thick layer of 5wt% Ir(ppy) 3 doped CBP and a 20 nm thick layer of 5wt% Ir(ppy) 3 doped TCTA. 3TPYMB and LiF/Al were vacuum deposited as the ETL and cathode , respectively . To investigate the role of UV O 3 treatment on PVP passivated metal oxides, AFM was used to examine the phase presence on PVP passivated NiO x , and XPS measurements were carried out to study the surface chemistry. To demonstrate that the PVP passivation is also beneficial to metal oxides for other optoelectronic applications, both perovskite solar cells with planar heterojunction (PHJ) structure and organic bulk heterojunction (BHJ) solar cells were prepared on different NiO x sam ples . PVP Passivation on NiO x HTLs and Its Application in OLEDs Devices To study the exciton quenching effect of NiO x HTL, EMLs composed of a 20 nm thick TCTA doped with 5wt% Ir(ppy) 3 were deposited onto as prepared NiO x , PVP passivated NiO x before and af ter UV O 3 treatment, and TAPC as a control HTL. Figure 4 1 A shows the PL spectra of all four samples with the data normalized to the PL intensity on top of the TAPC control sample . The as prepared NiO x sample shows the lowest PL intensity, indicating the strong luminescence quenching nature of the metal
77 oxide. With a thin PVP passivation layer, this quenching effect is effectively suppressed as evidenced by the similar PL intensity compared with the control sample. With subsequent UV O 3 trea tment on PVP passivated NiO x , the PL intensity is also very strong, indicating that the suppression of luminescence quenching is not significantly affected by the treatment. There are two possible exciton quenching mechanisms due to NiO x : (i) deep trap sta tes located at the NiO x surface act as the non radiative recombination sites , and (ii) energy/charge transfer from the emitter to the NiO x surface species. Since exciton quenching via mechanism (i) could only take place when the excitons reach the NiO x sur face, inserting a thin exciton blocking layer between the NiO x and EML should efficiently suppress this quenching effect . Based on this assumption, the author fabricated samples with 10 nm, 20 nm and 30 nm thick TAPC films as exciton blocking layers betwee n the NiO x film and the EML. The PL intensities of these samples are shown in Figure 4 1 B . Interestingly , with a 10 nm TAPC exciton blocking layer, the luminescence quenching is still very strong. By increasing the TAPC thickness, the luminescence quenchin g is gradually suppressed and the PL intensity is completely recovered with a TAPC thickness of 30 nm. These results indicate that there is a long range interaction between the quenching species at the NiO x surface and the excitons, leading to a quenching distance of ~30 nm. One possible candidate for such a quenching species is NiOOH , which has been reported to be a strong luminescence quencher. 78,126 The presence of the dipolar NiOOH species on NiO x surface is evidenced from Chapter 3. The strong NiOOH dipoles could facilitate the non radiative decay leading to a high exciton quenching rate at long distances. 127 From these data, it can be conclude d that the thin PVP passivation polymer plays two roles: (i) to suppress
78 the long range exci ton quenching of NiO x by passivating the NiOOH species; and (ii) to suppress the short range exciton quenching of NiO x by passivating its other surface defect s. As a result, the PL of the emitter on NiO x is effectively recovered with an ultra thin PVP pass ivation layer. To investigate the exciton quenching effects in OLEDs , devices were fabricated with f ive different HTLs: a 40 nm thick as prepared NiO x HTL, a 40 nm thick as prepared NiO x HTL followed by UV O 3 treatment ( UVO NiO x ), a 40 nm thick PVP passiva ted NiO x HTL ( P NiO x ), and a 40 nm thick PVP passivated NiO x HTL followed by UV O 3 treatment( P UVO NiO x ) , and a 40 nm thick TAPC as a reference . The energy band diagram is shown in Figure 4 2 A . [18,108] . The current density voltage ( J V ) curves of all four devices are shown in Figure 4 2 B , the corresponding luminescence voltage ( L V ) and current efficiency curves are shown in Figures 4 2 C and 4 2 D , respectively. The EL spectra of the OLEDs with NiO x HTLs are shown in Figure 4 3. The EL spectrum of the devices are measured using an Ocean Optics HR4000 spectrometer at the normal direction. The pixels are lighted at a constant current current density of 2.5 mA cm 2 . The good overlap of the spectrum implies that the emitting zone is c onfined around the same location in these devices. The device performance data are summarized in Table 4 1. The devices with the as prepared NiO x HTL, UVO NiO x HTL and P UVO NiO x HTL have simila r luminescence turn on voltage ( V on ~2.7 V ) as the control device with TAPC HTL , indicating a good energy level alignment between the HTL and EM L. And due to the superior hole transport properties of NiO x , the three abovementioned devices show higher current densities compared to the control device. The device wit h the as prepared NiO x HTL shows a very strong efficiency roll off, which can be attributed to
79 exciton quenching due to the as prepared NiO x HTL. With increasing bias, the emitting zone extends towards the HTL/EML interface, resulting in a stronger EL quen ching. With the UVO NiO x HTL, the device shows higher current densities with the maximum current efficiency shift ed to wards higher EL intensity , and these results are consistent with the data in Chapter 3. These results are attributed to the enhanced hole injection with more NiOOH dipolar species introduced onto the NiO x surface due to the UV O 3 treatment, 78 leading to the emitting zone extending further away from the NiO x surface and a reduction in efficiency roll off. However, due to the formation of more NiOOH species, UV O 3 treatment slightly aggravated the quenching effect of NiO x , as revealed by the lower PL intensity of the emitter on UVO NiO x as shown in Figure 4 4 , resulting in a lower maximum current efficiency. On the other hand, while PVP is a go od passivation layer, the device with the P NiO x HT L shows a very high V on ~5.0 V with a current efficiency less than 1 cd/A . This poor performance is attributed to the insulating PVP polymer on the surface of the P NiO x HTL, which prohibits hole injection from the NiO x HTL into the EML. On the contrary, the P UVO NiO x device shows the best performance with a highest maximum current efficiency of 90.8 + 2.1 Cd/A, which is ~20% higher than as prepared NiO x device, along with a significantly reduced efficiency roll off. The current efficiency curve of the P UVO NiO x device is also comparable to that of the TAPC control device due to the effectively suppressed exciton quenching. From these results, it can be concluded that UV O 3 treatment plays a significant role in improving the hole injection ability of the PVP passivated NiO x HTL. AFM and XPS of the PVP Passivated NiO x Films In order to study the effect of UV O 3 treatment on PVP passivated NiO x , AFM was use d to investigate the phases present on the surface of PVP passivated NiO x
80 before and after UV O 3 treatment. To minimize the surface contamination of NiO x , [90,128] all samples were stored in a vacuum chamber before measurements. The AFM topography images of 40 nm thick as prepared NiO x , P NiO x and P UVO NiO x films on top of ITO coated substrates are shown in Figure 4 5 A C . The as prepared NiO x film shows a RMS roughness of 1.8 nm with a maximum height variation of 10.4 nm. After PVP passivation, the film is flattened with an RMS of 0.8 nm and a maximum height variation of 5.1 nm. For the PVP passivated NiO x film followed by 5 min UV O 3 treatment, the film roughness slightly increases with an RMS of 1.1 nm and a maximum height variation of 5.9 nm. The correspo nding AFM phase images of all three samples are shown in Figure 4 5 D F . The phase image of the as prepared NiO x film shows a poly crystalline texture with an average grain size of 30 nm. The bright color in the NiO x texture completely disappears and the phase image is homogenous. T he dark O 3 treatment, a heterogeneous phase appears with both bright and dark regions. The decreased contrast in the phase image of P UVO NiO x film is also confirmed by the histogra m image of phase distribution as shown in Figure 4 6. These results indicate that PVP may react with the NiO x surface species upon UV O 3 treatment, yielding NiO x rich domains (bright) and PVP rich domains (dark). To investigate the change of the surface c ompositions, XPS was carried out to measure the binding energies of Ni 2p 3/2 , O 1s and C 1s. Interestingly, the Ni 2p 3/2 signal of the PVP passivated NiO x film remains the same after the UV O 3 treatment as shown in Figure 4 7 . Contradictory to our previous ly findings of UV O 3 treatment effects
81 on as prepared NiO x film, with a thin PVP passivation layer, there are no more higher oxidization states of Ni 3+ introduced after the UV O 3 treatment as revealed by the identical Ni 2p 3/2 signal spectra from both samples. Therefore , the enhancement in hole injection of PVP passivated NiO x is not due to the introduction of more dipolar NiOOH species as previously found in UV O 3 treated NiO x . To explain the significantly enhanced hole inject ion, a detail XPS study of the C 1s and O 1s signals was carried out to determine the surface chemical change s of the PVP passivated NiO x before and after UV O 3 treatment. Figure 4 8 A shows the C 1s spectrum of the P NiO x film. Based on the carbon configur ation in PVP, the C 1s spectrum is de convoluted into four carbon peaks corresponding to the signal associated with: (i) adventitious carbon with a binding energy (BE) of 285.0 e V (C1) ; (ii) the carbon linked to the carbonyl group with a BE of 285.4 eV (C2) ; (iii) the carbon nitrogen bond with a BE of 286.2 eV (C3) ; and (iv) the carbonyl bond with a BE of 287.8 eV (C4). 100,129 The ratio of the four carbons is consistent with the composition of the repeating unit of PVP (C1:C2:C3:C4=2:1:2:1) shown as an i nsert in Figure 4 8 A . Figure 4 8 B shows the O 1s spectrum of the P NiO x film. The spectrum is composed of a NiO x main peak at a BE of 529.5 eV, a NiO x defect peak at a BE of 531.2 eV, a carbonyl peak at a BE of 532.0 eV, and a surface absorbent peak at a B E of 533.2 eV. 100,129 From t he C 1s spectrum, it is determine d that the C=O double bond makes up 16.6 % of the whole C 1s spectrum, which is consistent with the value determined from the O 1s spectrum and the C:O atomic atio . Therefore, the C=O signals in the O 1s spectrum is entirely attributed to the carbonyl groups in PVP polymer. These results indicate that there is no chemical reaction between as -
82 prepared NiO x and PVP. Figure 4 8 C shows the C 1s spectrum of the P UVO NiO x film. After the UV O 3 treatment, the author observe s changes in the XPS spectrum at high binding energies due to the presence of the ester group with a BE of 288.9 eV. Figure 4 8 D shows the O 1s spectrum of the P UVO NiO x film where the author observe s similar changes in highe r binding energies due to the presence of the ester functional group with a BE of 532.9 eV. The carbon atoms from the carbonyl and ester groups makes up 20.9% and 14.8% of the C 1s spectrum, respectively, which are significantly larger than the correspondi ng values (11.8% and 8.4%) determined from the O 1s spectrum and the C:O atomic ratio. The discrepancy in the carbon composition determined from the C1s and O1s signals indicates the presence of additio nal oxygen atoms due to the formation of ester groups as a result of the chemical reaction between PVP and NiO x , indicating that NiO x shares its oxygen atoms with PVP to form carbonyl/ester groups after UV O 3 treatment . T he resulting chemical binding between PVP and NiO x surface species facilitates hole injection from P UVO NiO x into the adjacent EMLs, yielding devices with a high er efficien cy. The resolved details of the C 1s, O 1s and Ni 2p 3/2 XPS signals of the PVP passivated NiO x surface before ( P NiO x ) and after ( P UVO NiO x ) UV O 3 treatment are summarized in Table 4 2 . The compositions acquired are shown in the parentheses. The atomic ratios acquired are shown in the bottom row. The detailed analysis of the XPS data is described as follows: For the P NiO x sample, the O 1s signal from the carbonyl group makes up 49.8% of the entire O 1s spectrum and the rest of the oxygen comes from NiO x . And since the atomic ratio of O 1s to C 1s is 3:1 as shown in Table 2 , considering of the C 1s signal from the carbonyl group, which makes up 16.6% of the whole C 1s signals in the C 1s
83 spectrum, the calculation is consistent with the fact that 49.8% of the oxygen signal comes from the carbonyl group in PVP. Therefore, these results indicate that the entire O 1s spectrum is just the superimposition of O 1s from NiO x and O 1s from PVP, and there is no chemical reaction between NiO x and PVP before the UV O 3 treatment. For the P UVO NiO x , the O 1s signals from the C=O* bond and O=C O* bond make up 32.4% and 13.5% of the whole O 1s signals in the O 1s spectrum, respectively. However, since each ester group contributes equally one C=O* and one O=C O* bonds, the O 1s signals from the carbonyl group should make up 18.9% (32.4% 13.5%) of the entire O 1s spectrum, and the O 1s signals from the ester group make up 13.5 % of the entire O 1s spectrum. Considering of the atomic ratio of O 1s to C 1s, which is 1:1.6 in this case (shown in the bottom row of Table 4 2 ), it should be expect ed that the C 1s signals from the the carbonyl group and ester group to make up 11.8% (18 .9% divided by 1.6) and 8.4% (13.5% divided by 1.6) of the entire C 1s spectrum, respectively. However, these values are significantly smaller than their composition of 20.9% and 14.8% as determined from the C 1s spectrum. These results imply that the oxyg en atoms from lower binding energy regions (NiO x region) of the O 1s spectrum should also be attributing to the C=O* and O C=O* type bonding, and there is chemical bonding (oxygen sharing) between the NiO x surface species and PVP. It was also found that af ter UV O 3 treatment, most PVP polymer remained on the NiO x surface determined from our XPS analysis, further confirming our conclusion that instead of removing PVP, a short UV O 3 treatment on PVP passivated metal oxide surface facilitates chemical binding between the PVP passivation polymers and the NiO x surface species .
84 PVP Passivation on VO x HILs and Its Application in OLEDs Devices To illustrate the applicability of this approach to other metal oxides for OLEDs, VO x is used as another example and studied PVP passivation effect on its quenching mechanism. Here, the author measured the PL of a 20 nm thick TCTA:5wt% Ir(ppy) 3 film on VO x with and without a 10 nm thick of TAPC exciton blocking layer, and PVP passivated VO x film fo llowed by UV O 3 treatment. The PL intensities were normalized to that of the TAPC sample as shown in Figure 4 9 A . Unlike NiO x , VO x is an n type HIL with a deep electron affinity , 130 excitons are thus directly quenched at the VO x /EML interface as shown in t he figure. However, due to the absence of long distance (>10 nm) quenching process, a 10 nm thick TAPC layer on top of VO x is sufficient to suppress the luminescence quenching. The lower PL intensity from the sample with a PVP passivated VO x layer is due t o the thinner PVP passivation layer (< 5 nm), which cannot sufficiently suppress exciton quenching near the VO x surface based on a quenching mechanism different from that of NiO x . To study the passivation effect on OLEDs, devices were fabricated with the V O x HILs . Figure 4 9 B shows the J V L curves of the devices incorporating the as prepared VO x and PVP passivated VO x followed by the UV O 3 treatment ( P UVO VO x ), and Figure 4 9 C shows the current efficiency data. The device with P UVO VO x shows an improved current efficiency starting from high EL intensities (>2000 Cd m 2 ). However, the device does not show an enhanced current efficiency at low luminescence intensities. This could be explained by the absence of long range exciton quenching in the VO x devices. The EL is thus only quenched when emitting zone is close to VO x /EML interface, which occurs at relatively high luminescence.
85 PVP passivation on NiO x HTLs and Its Application in Solar Cell Devices To demonstrate that the PVP passivatio n is also beneficial to metal oxides for other optoelectronic applications, solar cells were fabricated on PVP passivated NiO x . The iodine (MAPbI 3 ) perovskite solar cells with P HJ were synthesized by the sequential deposition method. Lead Iodide (PbI 2 ) were first dissolved into dimethyl formamide (DMF) at a concentration of 400 mg mL 1 . The solution was kept at 80 Â°C and then spin casted onto the prepared substrates at a spin speed of 6000 rpm. The PbI 2 thickness was 150 + 10 nm. The methyl ammonium iodi de (MAI) solvent was prepared by dissolving MAI into isopropanol at a concentration of 10 mg mL 1 , and the solvent was kept at 90 Â°C in air. After PbI 2 deposition, the samples were immersed into the petri dishes filled with hot MAI solution for 1 min. Spon taneous reaction occurred and the film immediately turned from greenish yellow to dark brown due to the formation of MAPbI 3 . The petri dishes were then immersed into pure isopropanol solvent kept at room temperature to stop the reaction. The samples were t hen dried at 80 Â°C in glovebox for 30 min. a 30 nm thick (6,6) phenyl C61 butyric acid methyl ester (PC 60 BM) layer was spin coated onto the active layer and annealed at 80 Â°C for 30 min. The samples were then transferred into a thermal evaporator, where 15 nm thick fullerene (C 60 ), 5 nm thick 2,9 dimethyl 4,7 diphenyl 1,10 phenanthroline (BCP) and 130 nm sliver were subsequently vapor deposited. The XRD pattern and SEM images of the perovskite film formed on the P UVO NiO x are shown in Figure 4 10 . The J V curves (under AM 1.5G 1 sun illumination) of the PHJ Perovskite solar cells incorporating the a s prepared NiO x HTL, UVO NiO x HTL, P UVO NiO x HTL and PEDOT:PSS as a control device are shown in Figure 4 11 A . 131 The EQE spectrum of the PHJ perovskite solar cell incorporating P UVO NiO x HTL is shown in Figure 4 11 B . The device architecture is shown as an insert
86 in Figure 4 11 B . The device performance data are summarized in Table 4 3. The integrated J sc from the EQE spectr um is 20.24 mA cm 2 , which is consistent with the measured value of 20.27 mA cm 2 for the device with P UVO NiO x HTL. The PHJ perovskite solar cell with P UVO NiO x HTL shows the highest open circuit voltage (V oc ) of 1.04 + 0.02 V, the highest short circuit current density (J sc ) of 20.1 + 0. 4 mA cm 2 and the highest PCE of 10.9 + 0.3 % among all of the four samples, indicating the passivation technique is also beneficial for metal oxides used in solar cells . To demonstrate that the PVP passivation is also beneficial to metal oxides for OPVs , organic BHJ solar cells were fabricated on top of NiO x HTLs. The device architecture is shown in Figure 4 12 A . To prepare the NiO x HTLs, the diluted (0.1 M) nickel acetate precursor was spin coated onto the UV O 3 cleaned ITO substrates at 6000 rpm. A reduced annealing temperature of 300 Â°C was used to prevent the increase of ITO resistance at high temperatures. A longer UV O 3 treatment of 10 min was carried out on the PVP p assivated NiO x . This is due to the thicker PVP layer on low temperature processed NiO x surface with a stronger chemi sorption ability. 103 Poly(N 9Â´ heptadecanyl 2,7 carbazole alt 5,5 (4Â´,7Â´ di 2 thienyl 2Â´,1Â´,3Â´ benzothidiazole) (PCDTBT) and [6,6] phenyl C71 butyric acid methyl ester (PC 70 BM) were blended at a weight ratio of 1:4 and dissolved into chlorofo r m at a concentration of 20 mg/mL. After stirring in an inert atmosphere for 12 h, the solution was spin casted onto the NiO x HTLs or PEDOT:PSS HILs to form a 9 0 + 5 nm thick uniform film. 1 nm LiF and 100 nm Al were thermal evaporated as the cathode. Figure 4 12 B shows the J V curves of the BHJ solar cells incorporating the as prepared NiO x HTL, UVO NiO x HTL, P UVO NiO x HTL and PEDOT:PSS HIL under one sun illumination. The device
87 performance data are summarized in Table 4 4 . Serial resistances (R s e ) and shunt resistances (R sh ) were calculated by taking the inverse slopes at the V oc and J sc . The device with P UVO NiO x HTL shows a highest R sh and a relatively low R s e , however, the differences are not big enough to explain for its significantly improved V oc or J sc compared to the device with as prepared NiO x HTL or UVO NiO x HTL. The lower V oc of the device with as prepared NiO x HTL are also reported in other publications. 77 , 90 However, unlike the oxygen plasma treated NiO x device, our device with UVO NiO x shows even worse performance due to its lower J sc and fill factor (FF). Similar phenomena was also observed in our previous publication. 65 Since the NiO x films annealed at low temperatures have a lot of Ni(OH) 2 residue which can be easily turned into the dipolar species of NiOOH under O 3 exposure, the lo west J sc in those devcies could be attributed to the introduction of a significant amount of NiOOH species which quench excitons and hinder carrier extraction. The presence of NiOOH species was also confirmed by the darker color of UVO NiO x films, which, h owever, was not observed in the PVP passivated NiO x films even when the UV O 3 treatment time was increased from 5 min to 15 min. The results are consistent with our conclusion that the thin PVP passivation layer on top of NiO x films prevents further oxidiz ation of its surface hydroxide species to strong oxy hydroxide dipoles, which is detrimental to carrier dissociation and extraction. As a result of the PVP passivation, the OPV device with the P UVO NiO x HTL shows the highest V oc of 0.90 + 0.01 V and J sc o f 11.0 + 0.4 mA cm 2 , yielding a highest PCE of 5.5 + 0.2 % among all other three devices .
88 Table 4 1. OLED p erformance incorporating as prepared NiO x , UVO NiO x , P NiO x , and P UVO NiO x HTLs. Devices V on (V) V at 10 3 Cd m 2 (V) Maximum CE (Cd A 1 ) CE at 10 3 Cd m 2 (Cd A 1 ) CE at 10 4 Cd m 2 (Cd A 1 ) As prepared NiO x HTL 2.71 + 0.03 5.67 + 0.07 74.2 + 1.3 41.4 + 1.6 18.1 + 1.9 UVO NiO x HTL 2.65 + 0.02 4.29 + 0.10 71.0 + 2.2 66.5 + 2.0 36.9 + 2.1 P NiO x HTL 5.26 + 0.29 0.15 + 0.05 P UVO NiO x HTL 2.66 + 0.55 5.56 + 0.13 90.8 + 2.1 76.8 + 1.9 56.5 + 2.3 TAPC HTL (control) 2.66 + 0. 01 6 . 35 + 0. 08 9 2 . 0 + 0 . 9 80 . 2 + 1. 5 61 .5 + 2. 1
89 Table 4 2. The resolved XPS data of C1s, O1s and Ni 2p 3/2 signals of PVP passivated NiO x surface before and after UV O 3 treatment . C 1s Signal C1 (adv.) C2 C3 (C N) C4 (C=O) C5 (O C=O) P NiO x 285.0 (33.0%) 285.4 (17.8%) 286.2 (32.7%) 287.8 (16.6%) P UVO NiO x 285.0 (25.5%) 285.4 (13.3%) 286.2 (25.6%) 287.8 (20.9%) 288.9 (14.8%) O 1s Signal NiO (main) NiO (shoulder) C=O* O=C O* H 2 O, O 2 residue P NiO x 529.5 (28.5%) 531.2 (16.7%) 532.0 (49.8%) 533.3 (4.9%) P UVO NiO x 529.5 (26.9%) 531.1 (19.9%) 532.0 (32.4%) 532.9 (13.5%) 533.4 (7.3%) Ni 2p 3/2 Signal NiO (main) NiO (shoulder) NiO (shake up) as prepared NiO x a) 854.1 (20.6%) 856.0 (29.2%) 861.2 (50.3%) UVO NiO x b) 854.1 (16.3%) 855.9 (34.2%) 861.2 (49.5%) P NiO x 854.1 (24.6%) 856.0 (26.6%) 861.2 (48.8%) P UVO NiO x 854.1 (25.4%) 856.0 (28.3%) 861.2 (46.3%) Atomic Ratio Ni : O : C : N = 0.27 : 1 : 3 : 0.5 ( P NiO x ) Ni : O : C : N = 0.26 : 1 : 1.6 : 0.3 ( P UVO NiO x ) a,b) The Ni 2p 3/2 signals of as prepared and UV O 3 treated NiO x films are acquired from a previous XPS measurement and added for compare. The numbers outside the parenthesis indicate the binding energies (eV) of different components and the number inside the parenthesis indicate the component ratio. The aster isk s indicate the corresponding atoms.
90 Table 4 3. The performance of MAPbI 3 perovskite solar cells incorporating PEDOT:PSS, as prepared NiO x , UVO NiO x , and P UVO NiO x HTLs . Table 4 4. OPV performance incorporating PEDOT:PSS, as prepared NiO x , UVO NiO x , and P UVO NiO x HTLs. The standard deviation is reported in parenthesis . Devices of diff. HTLs V oc (V) J sc (mA/cm 2 ) FF (%) PCE (%) R s e cm 2 ) R sh cm 2 ) PEDOT:PSS 0.89 (0.004) 10.3 (0.2) 54.1 (0.3) 5.0 (0.1) 7.6 (1.0) 444.4 (72.7) as prepared NiO x 0.68 (0.006) 9.2 (0.2) 51.2 (0.2) 3.3 (0.1) 6.3 (0.9) 330.0 (50.5) UVO NiO x 0.86 (0.005) 7.6 (0.3) 42.1 (0.5) 2.8 (0.2) 9.4 (1.1) 262.5 (56.2) P UVO NiO x 0.90 (0.01) 11.0 (0.3) 54.7 (0.7) 5.5 (0.2) 6.7 (1.5) 518.1 (73.1)
91 Figure 4 1. The PL of a 20 nm TCTA: 5wt% Ir(ppy) 3 EML deposited onto a 40 nm NiO x or TAPC HTL . A ) The PL spectra on TAPC (black), as prepared NiO x (red), PVP passivated NiO x before (blue) and after (cyan) UV O 3 treatment. B ) The PL spectra on as prepared NiO x with 10 nm (blue), 20 nm (cyan) and 30 nm (pink) TAPC exciton blocking layers. The PL of the emitter on TAPC (black) and as prepared NiO x (red) are re plotted in B ) for compare . A B
92 Figure 4 2 . OLED s performance with TAPC or NiO x with different surface treatment. A ) Device structure and energy levels with respect to vacuum level of the materials used in thermal evaporated OLEDs. B ) J V characteristics of the OLEDs. C ) L V characteristics of the OLEDs. D ) current efficiency curves of the OLEDs with as prepared NiO x (black open square), UVO NiO x (black solid square), P NiO x (red open circle), P UVO NiO x ( red solid circle ), and TAPC control (blue dash) HTLs . A B C D
93 Figure 4 3. T he EL spectra of a phosphorescent green OLEDs incorporating the as prepared NiO x HTL, UVO NiO x HTL, and P UVO NiO x HTL . Figure 4 4. The PL of a 20 nm TCTA:5wt% Ir(ppy) 3 EML deposited onto 40 nm TAPC, as prepared NiO x , and UV O 3 treated NiO x .
94 Figure 4 5 . The surface morphology of 40 nm as prepared NiO x , P NiO x , and P UVO NiO x . A ) The AFM topography image of as prepared NiO x . B ) The AFM topography image of P NiO x . C) The AFM topography image of P UVO NiO x . D ) The AFM phase image of as prepared NiO x . E ) The AFM phase image of P NiO x . F ) The AFM phase image of P UVO NiO x . Insert of E ): the AFM phase image of P NiO x at a different scale bar . A C B D E F
95 Figure 4 6. The AFM phase distribution histogram of 40 nm as prepared NiO x , P NiO x , and P UVO NiO x HTLs .
96 Figure 4 7 . XPS spectra of the Ni 2p 3/2 signals from P NiO x and P UVO NiO x . A) The Ni 2p 3/2 signal from P NiO x . B) The Ni 2p 3/2 signal from P UVO NiO x . The two Ni 2p 3/2 signals are identical. A B
97 Figure 4 8 . XPS spectra of C 1s and O 1s signals from P NiO x and P UVO NiO x . A ) The C 1s signal from P NiO x (insert: the molecular structure of PVP with labelled C1, C2, C3 and C4) . B ) The O 1s signal from P NiO x . C ) The C 1s signal from P UVO NiO x . D ) The O 1s signal from P UVO NiO x . All of the XPS measurements were carried out at a take off angel of 45 Â°. A B C D
98 Figure 4 9 . OLED performance with as prepared and P UVO VO x HIL. A ) The PL of a 20 nm TCTA : 5wt% Ir(ppy) 3 EML deposited onto a 40 nm as prepared VO x (black solid), PVP passivated VO x after UV O 3 treatment (black dash), and as prepared VO x with 10 nm thick TAPC exciton blocking layer (grey solid). The base line is the PL intensity on 40 nm TAPC HTL. B ) The J V L c haracteristics of the OLEDs. C ) The current efficiency curves of the OLEDs with as prepared VO x (solid square) and P UVO VO x ( open circle ) HILs . A B C
99 Figure 4 10 . MAPbI 3 perovskite films formed on P UVO NiO x HTL. A) The XRD spectra of the perovskite film. B) The SEM images of the perovskite film. (b) A B
100 Figure 4 11 . The performance of perovskite solar cells with PEDOT:PSS or NiO x with different surface treatment. A) The J V curves of the perovskite solar cell s incorporating PEDOT:PSS (grey dash), as prepared NiO x HTL (black), UVO NiO x HTL (red) and P UVO NiO x HTL (blue) under AM 1.5 G one sun illumination. B ) The EQE spectrum of the perovskite solar cells incorporating P UVO NiO x HTLs. Inset: the device ar chitecture of the perovskite solar cells . B B
101 Figure 4 1 2 . The performance of OPVs with PEDOT:PSS or NiO x with different surface treatment. A ) Device architecture of the OPVs . B ) The J V curves of the devices with PEDOT:PSS (grey dash), as prepared NiO x HTL (black), UVO NiO x HTL (red) and P UVO NiO x HTL (blue) under AM 1.5 G one sun illumination. B A
102 CHAPTER 5 DEGRADATION STUDY OF SOLUTON PROCESSED SMALL MOLECULE PHOSPHORESCENT OLED Solution processed small molecule phosphorescent OLEDs have been demonstrated with high efficiencies in recent years due to the harvest of triplet excitons. However, in terms of the stability, they have substantially shorter operational lifetimes compared to their thermal evaporated counterparts. Photo induced degradation is proved to play a significant role causing the fast PL decay of the emitting materials. For example, cation radicals and oxidized species were found in small molecule phosphorescent emit ter during UV excitation, which is probably due to some photo oxidization process es . 132 However, in OLEDs, the degradation mechanism could be more sophisticated as electrical induced degradation also plays a role, and these two degradation mechanisms ca n h ave an impact on each other. 133 Time resolved PL and impedance spectroscopy (IS) measurements have been carried out to study the optical and electrical changes of the thermal evaporated OLEDs after degradation. 134 138 However, these studies alone were not able to tell the underlying mechanisms for the substantially different operational stabilities between thermal evaporated OLEDs and solution processed OLEDs. In this Chapter, the author presents a comprehensive study on the electrical induced and photo ind uced degradation of an optimized small molecule phosphorescent OLED structure with solution processed and thermal evaporated EMLs. Materials with high T g and high solubility are intentionally chosen in the system to eliminated morphology difference and the rmal induced degradation. By changing the dopant concentration in solution processed EML and using solvent with different impurity levels, the author was able to study how initial hole accumulation/hole trapping and solvent impurities within the EML affect s its degradation speed. Hole induced
103 degradation was studied by making single carrier devices, and excited states were introduced to those single carrier devices with UV excitation. It was found that the hole induced degradation of the EML is significantly accelerated with the presence of excited states. Time resolved PL was used to study the degradation of EML in OLEDs, and degradation scenarios were proposed to explain the fast EL decay in OLEDs with solution processed EMLs under two different circumstances. Experiment Section OLEDs are fabricated using PEDOT:PSS HILs and cross linkable PLEXCORE Â© hole transport layers (HTLs). 139,140 TPBi was used as the host as well as the hole blocking layer. Fo r thermal evaporated EML , Ir(ppy) 3 was doped into TPBi at a concentration of 10 wt%; for solution processed EML, fac tris(2 (3 p xylyl)phenyl)pyridine iridium(III) (TEG) was doped into TPBi at a concentration of 10 wt%, 20 wt% and 30 wt%. For solution pro cessed EMLs , TEG was used as the emitter due to its high solubility and similar opto electronic properties as Ir(ppy) 3 , 5 3 and toluene was used as the precursor solvent due to its inert non coordinating properties. 1 32 Two sources of toluene solvent are used to prepare the solution processed EML, one is fresh new and another has been dated in a N 2 filled glove box for over 18 months. Devices in this study were prepared on patterned ITO glass substrates. Before fabrication, the substrates were rinsed by sonica tion in acetone and isopropyl alcohol, followed by a 15 min UV O 3 treatment. A 40 nm thick PEDOT:PSS layer was spin nm thick cross linkable PLEXCORE Â© HTL was spin coated onto PEDOT:PSS HIL processed and thermal evaporated EMLs have a thickness of 35 nm, and the solution processed EML
104 2 filled glove box for complete drying. To make OLED devices, The samples were transferred into an evaporator chamber, where 10 nm TPBi as the hole blocking layer, 20 nm Alq 3 as the ETL , 1 nm LiF as the EIL and 100 nm Al as the cathode were deposited at a vacuum level of 1 Ã— 10 6 torr . To make the single carrier devices, hole only devices were prepared by depositing 10 nm molybdenum oxide (MoO x ) and 50 nm silver (Ag) as the counter electrode on top of the EML to block electron injection. The devices were encapsulated with a desiccant embedded cap glass by using an UV curable epoxy in a N 2 filled glove box. The active area of the devices was 4 mm 2 . Electrical characterizations were performed using a Keithley 2400 source meter and the EL intensity was recorded using a Keithl ey 6485 picoammeter with a silicon photodiode. The degradation curve was recorded in terms of EL drop and voltage rise with a time interval of 2 s. The EL spectra were measured with an Ocean Optics HR4000 spectrometer. The PL spectra and intensity were rec orded using a PerkinElmer LS 55 fluorescence spectrometer with a monochromatic excitation wavelength of 350 nm. Time resolved PL measurement was performed by focusing a pulsed nitrogen laser (the wavelength is 337 nm, Stanford Research Systems) onto the sa mple at an incident angle of 45 o . The time resolved PL was monitored by a Hamamatsu photomultiplier tube (PMT) with a 50 long pass filters were placed in front of PMT to block light with wavelength less than 500 nm . All measurements were carried out in ambient atmosphere and room temperature. Result and Discussion The structure of OLEDs used in the degradation study and the molecule structures of Ir(ppy) 3 , TEG and PLEXCORE Â© HTL are shown in Figure 5 1. [ 53,139 142 ]
105 Solution processed EMLs with three different TEG concentrations were prepared, corresponding to x = 10 wt%, 20 wt% and 30 wt%. For x = 30 wt%, both fresh and dated toluene solvent were used as the solvent precursor. OLED with thermal evaporated EML d oped with 10 wt% Ir(ppy) 3 was prepared as a control. Due to the large energy barrier between PLEXCORE Â© HTL and TPBi in HOMO , holes are directly injected into the Ir(ppy) 3 /TEG emitter and transport through the per colation of the dopants. 5 3, 1 4 1 Therefore, by changing the dopant concentration, the hole injection f rom HTL to EML and hole transport in the solution processed EMLs were able to be controlled . The J V L curves and current efficiency curves of the OLEDs are shown in Figure 5 2 A B . The EL spectra of the devices are shown in Figure 5 2 C . The grey dash corresponds to the thermal evaporated EML, the black, red, blue and green color corresponds to the solution processed EML with 10 wt% TEG, 20 wt% TEG, 30 wt% TEG, and 30 wt% TEG prepare d from dated toluene solvent, respectively. The device with 10 wt% TEG doped EML shows a larger V on compared to other devices. This result indicates that 10 wt% TEG in TPBi cannot form good hole percolation within the EML. With higher TEG concentration, th e device current increases accordingly due to improved hole injection and better hole transport in the EML. 142 OLED with solution processed EML doped with 20 wt% TEG already shows a higher current density than the device with thermal evaporated EML. With h igher TEG doping ratio, the device shows smaller efficiency roll off due to improved charge balance. However, the maximum current efficiencies remain the same with different TEG doping ratios, indicating the concentration quenching is not significant in th ose solution processed EMLs . 53 The operational lifetime of these devices were then tested under acceleration
106 conditions. 141 Figure 5 2 D shows the OLED degradation curves starting at an initial EL intensity of 10,000 Cd m 2 . EL decay curves are shown in top part of the figure and voltage rising curves are shown in bottom part of the figure. The luminance LT 50 lifetime (the time for the luminance to decay to 50 % of initial luminance) was recorded at an initial EL intensity of 20,000 Cd m 2 , 15,000 Cd m 2 , an d 10,000 Cd m 2 , respectively. And the acceleration factor (n) is fitted with the empirical equation as follows: Where t represents the LT 50 time, while L represents the starting EL intensity. The accelerated degradation curves and the fi tting is shown in Figure 5 3. The fitted acceleration factor is 1.83, which is consistent with the number of 1.8 reported in literature s . 141,142 The luminance LT 50 lifetime is 700 min for the OLED with thermal evaporated EML, corresponding to a LT 50 of 800 hours starting from 1000 Cd m 2 , which is comparable to the lifetime data reported with similar device structure. 143 Compared to the OLED with thermal evaporated EML, the four devices with solution processed EMLs have significantly shorter LT 50 life time . A mong them the device with 30 wt% TEG doped EML prepared from dated toluene solvent show the most rapid EL decay and voltage rising curves. Comparing it with the device in which the 30 wt% TEG doped EML was prepared from fresh toluene, the J V L and efficiency curves are very similar to those of the other device. Thus it is most likely that the significantly shorter operational lifetime here should be attributed to solvent impurities. The voltage rising and EL decay curves are symmetric in OLEDs with thermal evaporated EML and solution processed EMLs with 20 wt% and 30 wt% TEG concentration. However, in the device with 10 wt% TEG
107 doped EML, the EL drops rapidly while the voltage rising is not significant. This result indicates that for solution process ed OLED with 10% TEG, the degradation mechanism may have a different profile as in other devices. To investigate the electrical induced degradation, hole only devices of the abovementioned EMLs were prepared and operated under constant current in dark cond ition. Figure 5 4 A shows the hole current of these devices under direct current (DC) bias. Same color was used to label the devices incorporating thermal evaporated EML and solution processed EML for consistency purpose. The hole current is enhanced with h igher TEG concentrations as expected. Figure 5 4 B shows the bias voltage curves of the hole only devices under a constant current of 2.5 mA cm 2 . The device with thermal evaporated EML does not show any voltage ris e after being driven for more than 10 hour s, while all of the devices with solution processed EMLs show a fast voltage ris e within the first several minutes, then followed by a linear and more gradual voltage rise with time. A significantly faster voltage rise is found in the hole only device with 10 wt% TEG. Since the EMLs with 20 wt% and 30 wt% TEG (with fresh toluene) show very similar degradation in both OLED s and hole only devices, the data for the devices with 20% TEG is not presented in the following degradation studies. To accelerate the ho le induced degradation, the devices were operated at a higher current of 25 mA cm 2 for 1 hour, and an in situ PL measurement was carried out to study the optical change of the EMLs. The voltage rising curves are shown in figure 5 4 C , the PL of the corresponding EMLs are shown in Figure 5 5 A D . Each plot has 4 PL spectra the solid/dash line s represent the PL when device was not/was under electrical driving. The black/red lines represent the PL before/after the device was
108 continuousl y driven for 1 hour. Due to the triplet polaron annihilation (TPA) process, 144 147 all of the EMLs show a drop in PL when the device was electrically driven as revealed by the discrepancy of solid and dash lines. For the hole only device with thermal evapo rated EML, the PL spectra before and after 1 hour of continuous driving overlap with each other. The unchanged PL intensity as well as the stable operating voltage under electrical driving reveal that hole induced degradation is not likely to occur in ther mal evaporated EML. For the hole only devices with solution processed EML, however, the PL does not fully recover to its initial status after 1 hour of continuous driving, indicating that hole induced degradation occur s in the solution processed EML. The h ole only device with 10 wt% TEG doped EML shows the biggest PL drop . C ombined with its rapid voltage rise under electrical driving, it indicates that the hole induced degradation play a more significant role in 10 wt% TEG doped EML than in other EMLs. As t he 10 wt% TEG cannot form good hole percolation, a large amount of holes are trapped at the HTL/EML interface and in the bulk of EML at the beginning of electrical driving . These trapped holes could trigger the cation radical formation process that is detr imental to the chemical stability of the surrounding materials. 132,133,138 Although hole induced degradation was observed in solution processed EML, the degradation is not very significant in solution processed EML with 20 wt% and 30 wt% TEG due to improve d hole injection and transport. 1 4 1 Nevertheless, a faster rising rate in voltage and bigger drop in PL due to hole induced degradation were observed in the hole only device prepared from dated toluene solvent. This could be attributed to presence of reacti ve solvent impurities (for example, singlet oxygen) in dated toluene
109 that accelerate the hole induced degradation of the EML . 132, 148 To verify this, gas chromatography mass spectrometry (GC MS) measurement was carried out to measure the solvent impurity sp ecies in both fresh and dated toluene solvent. The spectrum in a 12 min 20 min time scale window are shown in Figure 5 6 . The three peaks corresponding to 12.66 min, 15.78 min and 18.60 min in both fresh and dated toluene are identified as siloxane species from the GC column. However, in dated toluene solvent, an additional peak at 16.52 min was observed, which is identified as benzaldehyde. It is known that methyl substituted aromatic system can be oxidized by singlet oxygen to form aromatic aldehydes. 132, 149 Thus the existence of benzaldehyde in dated toluene solvent indicates that there are oxygen trapped in the solvent. Since trace amount of oxygen is known to have a pronounced effect on the chemical stability of the organic complexes, 132 the faster hole induced degradation in the hole only device prepared from dated toluene should be attributed to the trapped oxygen in the bulk EML from the dated solvent . The effect of excited states on hole induced degradation of solution processed EML was studied. The photo excitation was performed on the hole only devices using a UV lamp (Spectroline EA 140) during the first 30 min of electrical driving. The PL of the pixel area is corresponding to an intensity of 600 Cd m 2 . After 30 min, the UV lamp was turned off, and the hole only devices were continuously driven under the dark condition for another 30 min. The voltage rise in the device where the 30 wt% TEG EML was prepared from dated toluene solvent was recorded and shown in Figure 5 7 (red line) . T he voltage ris e in dark condition is also plotted in Figure 5 7 for compare purpose (black line). With the on and off of the UV lamp, the voltage r ise curve has two distinct
110 stages: During the first stage, the UV lamp is on and hole induced degradation is accompanied by the excited states of EML. The linear voltage rising slope is larger than that of the reference curve . D uring the second stage, the UV lamp is off and the excited states do not exist . T he driving voltage jumps upward immediately to compensate for the loss of photo induced current. The fast voltage rise at the beginning of the second stage is probably due to the filling of shallow hole traps which is not occupied in the first stage due to the UV excitation. Nevertheless, the voltage rise is significantly sl ower in the second stage and the rate of the rise in the linear region is similar to the reference curve. These results reveal that the hole induced degradation of EML could be accelerated by the excited states . It has been reported that the excited states of trace oxygen impurities in EML could drastically speed up the chemical degradation process. 132 This could explain why in OLEDs, the EMLs prepared from fresh and dated toluene show such a big difference in voltage ris e /EL decay rate, while in hole only devices, the difference becomes significantly smaller. It still needs to be answered that why OLED with 10 wt% TEG doped EML shows significantly smaller voltage ris e compared to the hole only device, and why the smaller voltage rise is corresponding to a s tronger EL decay. Before answering these questions, the overall degradation of EML in OLEDs was investigated using time resolved PL measurement. The degradation was performed in the same condition as shown in Figure 5 2 D . After driving for 25 min, the time resolved PL intensities of four OLEDs incorporating thermal evaporated EML, solution processed EML with 10 wt% TEG, 30 wt% TEG and 30 wt% TEG prepared from dated toluene solvent are shown in Figure 5 8 A D . Th e black and red color represents the devices be fore and after degradation
111 respectively . All of the four devices show reduced exciton lifetime of the phosphorescent emitter after degradation, indicating more exciton quenching channels in the EML after degradation. The reduction of exciton lifetime is mo st significant in the device prepared from dated toluene solvent, indicating most severely degradation in the EML. In the OLED with 10 wt% TEG, however, the reduction of the exciton lifetime is comparable to that in the other devices, even though the devic e operational lifetime is much shorter in this device. Figure 5 9 shows the PL intensity versus EL intensity of all of the devices after degradation. The obvious smaller PL decay of the OLED with 10% TEG doped EML implies that the degradation should occur within a very small part of the EML. For the rest of the devices, the PL intensity is proportional to the EL intensity with a constant ratio even after degradation, as evidenced by the linear fitting. (PL is normalized to 1 at t=0 when EL = 10,000 Cd m 2 ). Therefore, the EL decay of these devices should be correlated with the bulk EML degradation. To explain the mechanisms for the fast degradation of OLEDs with solution processed EMLs, two scenarios should be considered as shown in Figure 5 10 . Scenario A Interfacial degradation of the EML: For the device with 10 wt% TEG doped EML, the hole injection/transport is problematic, thus the fast voltage rise in the hole only device is attribute to the holes accumulated at the HTL/EML interface and the entire EML . In the OLED, however, due to the presence of electron current and severe charge imbalance, emitting zone is confined closely at the HTL/EML interface where most of the degradation takes place, while the rest of the EML is unaffected due to the absence of holes or excited states. Therefore, although the EL decay is fast, the voltage
112 rise and overall EML degradation is not significant in the solution processed OLED with 10 wt% TEG. Scenario B: Degradation of the bulk EML. For the devices with 20 wt% and 30 wt% TEG doped EML, the hole injection/transport is significantly improved. However, hole induced degradation is still observed in those solution processed EMLs, and this degradation process is further accelerated with the presence of excited states in the OLEDs. The correlation between the EL and PL decays in these OLEDs indicates that compared to the device with thermal evaporated EML, the inferior operational stabilities of the devices with solution processed EMLs should be attributed to the rapid degrad ation in the bulk of the solution processed EMLs. The degradation rate is significantly affected by the impurity level of the precursor solvent. In summary, the author demonstrated a system for the comprehensive study of the degradation mechanism of OLEDs with solution processed EMLs. Evidence is found that solution processed EML is more prone to hole induced degradation as compared to the thermal evaporated EML. The degradation speed depends on the amount of initial hole accumulati on/traps and solvent impu rities. T he degradation process is further accelerated with the presence of excited states. Follow up study reveals that for the solution processed OLED with good hole injection/transport properties, the operational stability of the device depends on the d egr adation rate of the overall EML. S olvent impurities within the EML could play the most significant role in its short operational lifetime.
113 Figure 5 1 . OLED architecture and molecule structure. A ) The energy diagram of OLEDs with thermal evaporated EML . B ) The energy diagram of OLEDs with solution processed EML. C) Molecule structure of emitters and HTL . A B C
114 Figure 5 2 . Performance and operational stability of the OLEDs with solution processed or thermal evaporated EMLs. A ) J V L curves. B ) Current efficiency curves. C ) EL spectra . D ) the EL decay and voltage rising curves of the OLEDs driving at an initial EL of 10,000 C d m 2 . The grey color corresponds to the thermal evaporated EML, the black, red, blue and green color corresponds to the solution processed EML with 10 wt% TEG, 20 wt% TEG, 30 wt% TEG, and 30 wt% TEG with dated toluene solvent respectively . A B C D
115 Figure 5 3 . The accelerated degradation model. A ) The EL degradation starting from 20,000 Cd m 2 , B) The EL degradation starting from 15,000 Cd m 2 . C ) The EL degradation starting from 10,000 Cd m 2 . D ) The fitted data yield an acceleration factor of 1.83. A B C D
116 Figure 5 4 . Hole only devices with thermal evaporated or solution processed EMLs. A) current density of the hole only devices ; B ) Voltage rising curves at a constant current density of 2.5 mA cm 2 ; C ) Voltage rising curves at a constant current density of 25 mA cm 2 . A B C
117 Figure 5 5 . PL degradation of hole only devices with thermal evaporated or solution processed EMLS. A ) PL of thermal evaporated EML . B ) PL of solution processed EML with 10 wt% TEG . C ) PL of solution processed EML with 30 wt% TEG . D ) PL of solution processed EML with 30 wt% TEG prepared from dated toluene solvent. Sold (dash) line represent PL when device was not (was) electrical driven, black (red) color represent PL before (after) 1 hour of continuous driving. All of the PL spectra were normalized to the peak intensity of the black solid line. The excitation wavelength is 350 nm. A B C D
118 Figure 5 6 . GC MS spectrum of both fresh toluene solvent and dated toluene solvent. The molecule structure of b enzaldehyde is shown in the spectrum of dated toluene solvent.
119 Figure 5 7 . Voltage ris e curves of hole only devices with or without UV excitation. The EML is 30 wt% TEG doped TPBi prepared from dated toluene solvent. Red line: the hole only device was excited with a UV lamp during first 30 minutes driving ; Black line: the hole only device driven under dark condition.
120 Figure 5 8 . Time resolved PL of th e OLEDs before and after degradation. A ) Time resolved PL of thermal evaporated EML . B ) Time resolved PL of solution processed EML with 10 wt% TEG . C ) Time resolved PL of solution processed EML with 30 wt% TEG . D ) Time resolved PL of solution processed EML with 30 wt% TEG prepared from dated toluene solvent. Black (red) line represents the time resolved PL of the device before (after) degradation. A B C D
121 Figure 5 9 . The PL EL intensity of the OLEDs with different EMLs after continuous driving for 25 min.
122 Figure 5 10 . The two schematic degradation mechanism s for solution processed EMLs in hole only devices and OLED devices.
123 LIST OF REFERENCES (1) Bernanose, A.; Comte, M.; Vouaux, P. J. Chem. Phys. 1953 , 50 , 64. (2) Bernanose, A.; Vouaux, P. J. Chem. Phys . 1953 , 50 , 261. (3) Bernanose, A. J. Chem. Phys. 1955 , 52 , 396. (4) Bernanose, A.; Vouaux, P. J. Chem. Phys. 1955 , 52 , 509. (5) Kallmann, H.; Pope, M. J. Chem. Phys. 1960 , 32 , 300. (6) Kallmann, H.; Pope, M. Nature 1960 , 186 , 4718. (7) Mark, P.; Helfrich, W. J. Appl. Phys. 1962 , 33 , 205. (8) Pope, M.; Kallmann, H. P.; Magnante, P. J. Chem. Phys. 1963 , 38 , 2042. (9) Helfrich, W.; Schneider, W. Phys. Rev. Lett. 1965 , 14 , 229. (10) Tang, C. W.; Vanslyke, S. A. Appl. Phys. Lett. 1987 , 51 , 913. (11) Seo, J. H.; Jin Y.; Brzezinski, J. Z.; Walker, B.; N guyen, T. Q. Chem. Phys. Chem. 2009 , 10 , 1023 1027. (12) Tsutsui, T.; Aminaka, E.; Lin, C. P.; Kim, D. U. Phil. Trans. Soc. A 1997 , 355 , 801. (13) Chavoor, G.; Jin, X.; Ellaboudy, A.; Fu, X. X.; Kang, X. N.; Zhang, B.; Zhang, G. Y. Proc. of SPIE , 2011 , 8123 . (14) Park, J . W.; Lee, G. H.; Kwon, Y. Y.; Park, K. W.; Lee, J.; Jin, Y. W.; Nah, Y. C.; Kim, H. Org. Electron. 2014 , 15 , 2178 2183. (15) Koo, W. H.; Jeong, S. M.; Araoka, F.; Ishikawa, K.; Nishimura, S.; Toyooka, T.; Takezoe, H. Nat. Photonics 2010 , 4 , 222 226. (16) Lee, Y. J.; Kim, S. H.; Huh, J.; Kim, G. H.; Lee, Y. H.; Cho, S. H.; Kim, Y. C.; Do, Y. R. Appl. Phys. Lett. 2003 , 82 , 3779. (17) Cho, T. Y.; Lin, C. L.; Wu, C. C.; Appl. Phys. Lett. 2006 , DOI: 10.1063/1.2185077. (18) Xiang, C.; Koo, W.; So, F.; Sasabe, H.; Kido, J. Light Sci. Appl. 2013 , 2 , e74; doi:10.1038/lsa.2013.30 (19) Ou, Q. D.; Zhou, L.; Li, Y. Q.; Shen, S.; Chen, J. D. Li, C.; Wang, Q. K.; Lee, S. T.; Tang, J. X. Adv. Funct . Mater. 2014 , 24 , 7392.
124 (20) Pardo, D. A.; Jabbour, G. E.; Peyghambarian, N.; Adv. Mater. 2000 , 12 , 1249. (21) Subramanian, V. presented at Solid State Dev. Res. Conf. Edinburgh, UK, September, 2008 . (22) Han, T. H.; Lee, Y.; Choi, M. R.; Woo, S. H.; Bae, S. H.; Hong, B. H.; Ahn, J. H.; Lee, T. W . Nat . Photonics 2012 , 6 , 105 110. (23) Gevaux, D. Nat. Photonics 200 7 , 1 , 567 568. (24) Hecht, D. S.; Hu, L.; Irvin, G. Adv. Mater. 2011 , 23 , 1482 1513. (25) Cairns, D. R.; Witte, R. P.; Sparacin, D. K.; Sachsman, S. M.; Paine, D. C.; Crawford, G. P.; Newton, R. R. Appl. Phys. Lett. 2000 , 76 , 1425 1427. (26) Krebs, F.C.; Jorgensen, M.; N orrman, K.; Hagemann, O.; Alstrup, J.; Nielsen, T. D.; Fyenbo, J.; Larsen, K.; Kristensen, J. Sol. Energy Mater. Sol. Cells 2009 , 93 , 422 441. (27) Krebs, F. C. Org. Electron. 2009 , 10 , 761 768. (28) Kim,A.; Won, Y.; Woo, K.; Jeong, S.; Moon, J. Adv. Funct. Mater. 2 014 , 24 , 2462 2471. (29) Vosgueritchian, M.; Lipomi, D. J.; Bao, Z. Adv. Funct. Mater. 2012 , 22, 421 428. (30) Granlund, T.; Pettersson, L. A. A.; InganÃ¤s, O. J. Appl. Phys. 2001 , 89 , 5897. (31) Ghosh, S.; InganÃ¤s, O. Synth. Met. 2001 , 121 , 1321. (32) Jonas, F.; Karbach, A.; Muys, B.; Thillo, E.; Wehrmann, R.; Elschner, A.; Dujardin, R. U.S. Patent No. 6,083,635 ( 2000 ). (33) Kim, J. Y.; Jung, J. H.; Lee, D. E.; Joo, J. Synth. Met. 2002 , 126 , 311. (34) Zhang, F.; Johansson, M.; Andersson, M. R.; Hummelen, J. C.; InganÃ¤s, O. Adv. Mater. 2002 , 14 , 662. (35) Kushto, G. P.; Kim, W. Kafafi, Z. H. Appl. Phys. Lett. 2005 , 86 , 093502. (36) Hu, L.; Kim, H. S.; Lee, J. Y.; Peumans, P.; Cui, Y. ACS Nano. 2010 , 4 , 2955 2963. (37) Lee, J. Y.; Connor, S. T.; Cui, Y.; Peumans, P. Nano. Lett . 2008 , 8 , 689 692. (38) Zeng, X . Y.; Zhang, Q. K.; Yu, R. M. Lu, C. Z. Adv. Mater. 2010 , 22 , 4484 4488. (39) Gaynor, W.; Burkhard, G. F.; McGehee, M. D.; Peumans, P. Adv. Mater. 2011 , 23 , 2905 2910.
125 (40) Li, L.; Yu, Z.; Hu, W.; Chang, C. H.; Chen, Q.; Pei, Q. Adv. Mater. 2011 , 23 , 5563 5567. (41) Yu, Z.; Zhang, Q.; Li, L.; Chen, Q.; Niu, X.; Liu J.; Pei, Q. Adv. Mater. 2011 , 23 , 664 668. (42) Ahmed, E.; Earmme, T.; Jenekhe, S. A. Adv. Funct. Mater. 2011 , 21 , 3889 3899. (43) Earmme, T.; Jenekhe, S. A. J. Mater. Chem. 2012 , 22 , 4660. (44) Earmme, T.; Ahmed, E.; Jenekhe , S. A. Adv. Mater. 2010, 22 , 4744 4748. (45) Earmme, T.; Jenekhe, S. A. Adv. Funct. Mater. 2012 , 22 , 5126 5136. (46) Earmme, T.; Jenekhe, S. A. J. Mater. Chem. 2012 , 22 , 4660 4668. (47) Huang, F.; Shih, P. I.; Shu, C. F.; Chi, Y.; Jen, A. K. Y. Adv. Mater. 2009 , 21 , 361 365. (48) Ye, T.; Shao, S.; Chen, J.; Wang, L.; Ma, D. ACS Appl. Mater. Interfaces 2011 , 3 , 410 416. (49) Yang, X.; MÃ¼ller, D. C.; Neher, D.; Meerholz, K. Adv. Mater. 2006 , 18 , 948 954. (50) Niu, Y. H.; Liu, M. S.; Ka, J. W.; Jen, A. K. Y. Appl. Phys. Lett . 2006 , 88 , 093505. (51) Niu, Y. H.; Liu, M. S.; Ka, J. W.; Bardeker, J.; Zin, M. T.; Schofield, R.; Chi, Y.; Jen A. K. Y. Adv. Mater. 2007 , 19 , 300 304. (52) Zacharias, P.; Gather, M. C.; Rojahn, M.; Nuyken, O.; Meerholz, K. Angew. Chem. Int. Ed. 2007 , 46 , 4388 4392. (53) Rehmann, N.; Hertel, D.; Meerholz, K.; Becker, H.; Heun, S. Appl. Phys. Lett. 2007 , 91 , 103507. (54) Cheng, Y. J.; Liu, M. S.; Zhang, Y.; Niu, Y.; Huang, F.; Ka, J. W.; Yip, H. L.; Tian, Y.; Jen, A. K. Y. Chem. Mater. 2008 , 20 , 413 422. (55) Ma, B.; Lauterwasser, F.; Deng, L.; Zonte, C. S.; Kim, B. J.; FrÃ©chet, J. M. J. Chem. Mater. 2007 , 19 , 4827 4832. (56) Lim, Y.; Park, Y. S.; Kang, Y.; Jang, D. Y.; Kim, J. H.; Kim, J. J.; Sellinger, A.; Yoon, D. Y. J. Am . Chem. Soc. 2011 , 133 , 1375 1382. (57) Liaptsis, G.; Hertel, D.; Meerholz K. Angew. Chem. Int. Ed. 2013 , 52 , 9563 9567. (58) Aizawa, N.; Pu, Y. J.; Chiba, T.; Kawata, S.; Sasabe, H.; Kido, J. Adv. Mater. 2014 , 26 , 7543 7546.
126 (59) Jiang, W.; Ban, X.; Ye, M.; Sun, Y.; Duan L.; Qiu, Y. J. Mater. Chem. C , 2015 , 3 , 243. (60) Park, J.; Lee, C.; Jung, J.; Kang, H.; Kim, K. H.; Ma, B.; Kim, B. J. Adv. Funct. Mater. 2014 , 24 , 7588 7596. (61) Lee, J.; Han, H.; Lee, J.; Yoon S. C.; Lee, C. J. Mater. Chem. C 2014 , 2 , 1474. (62) Deng, L.; Furuta, P. T.; Garon, S.; Li, J.; Kavulak, D.; Thompson, M. E.; FrÃ©chet, J. M. J. Chem. Mater . 2006 , 18 , 386 395. (63) W. M.; Lademann, H.; MÃ¼ller, D. C.; Meerholz, K. Adv. Mater . 2009 , 21 , 879 884. (64) Yan, H.; Scott , B. J.; Huang, Q.; Marks, T. J. Adv. Mater. 2004 , 16 , 1948 1953. (65) Manders, J. R.; Tsang, S. W.; Hartel, M. J.; Lai, T. H.; Chen, S.; Amb, C. M.; Reynolds, J. R.; So, F. Adv. Funct. Mater. 2013 , 23 , 2993 3001. (66) Lee, T. W.; Noh, T.; Shin, H. W.; Kwon, O.; Par k, J. J.; Choi, B. K.; Kim, M. S.; Shin, D. W.; Kim, Y. R. Adv. Funct. Mater. 2009 , 19 , 1625 1630. (67) Xing, X.; Zhong, L.; Zhang, L.; Chen, Z.; Qu, B.; Chen, E.; Xiao, L.; Gong, Q.; J. Phys. Chem . C 2013 , 117 , 25405 25408. (68) So, F.; Kondakov, D. Y. Adv. Mater. 2010 , 22 , 3762 3777. (69) Smith, A. R. G.; Ruggles, J. L.; Cavaye, H.; Shaw, P. E.; Darwish, T. A.; James, M.; Gentle, I. R.; Burn, P. L. Adv. Funct. Mater. 2011 , 21 , 2225 2231. (70) Hu, W. Niu, X.; Li, L. Yun, S.; Yu, Z.; Pei, Q. Nanotechnology 2012 , 23 , 344002. (71) M adaria, A. R.; Kumar, A.; Ishikawa, F. N.; Zhou, C. Nano. Res. 2010 , 3 , 564 573. (72) Madaria, A. R.; Kumar, A.; Zhou, C. Nanotechnology 2011 , 22 , 245201. (73) Reinhard, M.; Eckstein, R.; Slobodskyy, A.; Lemmer, U.; Colsmann, A. Org. Electron. 2013 , 14 , 273 277. (74) Barrera, D.; Lee, Y. J.; Hsu, J. W. P. Sol. Energy Mater. Sol. Cells 2014 , 125 , 27 32. (75) Jasieniak, J. J.; Seifter, J.; Jo, J.; Mates, T.; Heeger, A. J. Adv. Funct. Mater. 2012 , 22 , 2594 2605. (76) Li, B.; Ren, H.; Yuan, H.; Karim, A.; Gong, X. ACS Photonics 2014 , 1 , 87 90.
127 (77) Steirer, K. X.; Ndione, P. F.; Widjonarko, N. Edwin; Lloyd, M. T.; Meyer, J.; Ratcliff, E. L.; Kahn, A.; Armstrong, N. R.; Curtis, C. J.; Ginley, D. S.; Berry, J. J.; Olson, D. C. Adv. Energy Mater. 2011 , 1 , 813 820. (78) Ratcliff, E. L.; Meyer, J.; Steirer, K. X.; Garcia, A.; Berry, J. J.; Ginley, D. S.; Olson, D. C.; Kahn, A.; Armstrong, N. R. Chem. Mater. 2011 , 23 , 4988 5000. (79) Ratcliff, E. L.; Bakus II, R. C.; Welch, G. C.; van der Poll, T. S.; Garcia, A.; Cowan, S. R.; Macleod, B. A.; Ginley,D. S. ; Bazan, G. C.; Olson, D. C. J. Mater. Chem. C 2013 , 1 , 6223 6234. (80) Wong, K. H.; Ananthanarayanan, K.; Heinemann, M. D.; Luther, J.; Balaya, P. Sol. Energy 2012 , 86 , 3190 3195. (81) Betancur, R.; MaymÃ³, M.; Elias, X.; Vuong, L. T.; Martorell, J. Sol. Energy Mate r. Sol. Cells 2011, 95 , 735. (82) Adler, D.; Feinleib, J. Phys. Rev. B 1970 , 2 , 3112. (83) HÃ¼fner, S.; Steiner, P.; Sander, I.; Reinert, F.; Schmitt, H.; Neumann, M.; Witzel, S. Solid State Commun . 1991 , 80 , 869. (84) HÃ¼fner, S.; Osterwalder, J.; Riesterer, T.; Hulliger, F. Solid State Commun. 1984 , 52 , 793. (85) Juybari, H. A.; Bagheri Mohagheghi, M. M.; Shokooh Saremi, M. J. Alloys Compd. 2011 , 509, 2770. (86) Sawatzky, G. A.; Allen, J. W. Phys. Rev. Lett. 1984 , 53 , 2339. (87) KrÃ¶ger, F. A.; J. Phys. Chem. Solids 1968 , 29 , 1889. (88) Mrowec, S.; Grzesik, Z. J. Phys. Chem. Solids 2004 , 65 , 1651. (89) Ratcliff, E. L.; Meyer, J.; Steirer, K. X.; Armstrong, N. R.; Olson, D.; Kahn, A. Org. Electron. 2012 , 13 , 744. (90) Steirer, K. X.; Chesin, J. P.; Widjonarko, N. E.; Berry, J. J.; Miedaner, A. D.; Ginley, S.; Olson, D. C. Org. Electron. 2010 , 11 , 1414 1418. (91) Jung, J.; Kim, D. L.; Oh, S. H.; Kim, H. J. Sol. Energy Mater. Sol. Cells 2012 , 102 , 103 108. (92) Chan, I. M.; Hsu, T. Y.; Hong, F. C. Appl. Phys. Lett. 2002 , 81 , 1899 1901. (93) Chan, I. M.; Hong, F. C. Thin Solid Films 2004 , 450 , 304 311. (94) Im, H. C.; Choo, D. C.; Kim, T.W.; Kim, J. H.; Seo, J. H.; Kim, Y. K. Thin Solid Films 2007 , 515 , 5099 6102.
128 (95) Zilberberg, K.; Trost, S.; Meyer, J.; Kahn, A.; Behrendt, A.; LÃ¼tzenkirchen Hecht, D.; Frahm, R.; Riedl, T. Ad v. Funct. Mater . 2011 , 21 , 4776 4783. (96) Meyer, J.; Zilberberg, K.; Riedl, T.; Kahn, A. J. Appl. Phys. 2011 , 110 , 033710. (97) Zilberberg, K.; Trost, S.; Schmidt, H.; Riedl, T. Adv. Energy Mater. 2011 , 1 , 377 381. (98) Park, Y. R.; Kim, K. J. J. Crystal Growth . 2003 , 2 58 , 380 384. (99) Chakroune, N.; Viau, G.; Ammar, S.; Poul, L.; Veautier, D.; Chehimi, M. M.; Mangeney, C.; Villain, F.; FiÃ©vet, F. Langmuir , 2005 , 21 , 6788 6796. (100) Payne (101) Koo, Y. M.; Choi, S. J.; Chu, T. Y.; Song, O. K.; Shin, W. J.; Lee, J. Y.; Kim, J. C.; Yoon, T. H.; J. Appl. Phys. 2008 , 104 , 123707. (102) Small, C. E.; Tsang, S. W.; Kido, J.; So, S. K.; So, F. Adv. Funct. Mater. 2012 , 22 , 3261 3266. (103) GÃ¶pel, W. Surf. Sci. 1977 , 62 , 165. (104) Langell, M. A.; Berrie, C. L.; Nassir M. H.; Wulser, K. W. Surf. Sci. 1994 , 320 , 25 38. (105) Langell M. A.; Nassir, M. H. J. Phys.Chem. 1995 , 99 , 4162 4169. (106) Roberts M. W.; Smart, R. S. C. Surf. Sci. 1980 , 100 , 590 604. (107) Jiang, J. Wang, X.; Zhang, Q.; L i, J.; Zhang, X. X. Phys. Chem. Chem. Phys. 2013 , 15 , 6875. (108) Cai, M.; Xiao, T.; Hellerich, E.; Chen, Y.; Shinar, R.; Shinar, J. Adv. Mater. 2011 , 23 , 3590 3596. (109) Kabra, D.; Song, M. H.; Wenger, B.; Friend, R. H.; Snaith, H. J. Adv. Mater. 2008 , 20 , 3447 3452 . (110) Bolink, H. J.; Coronado, E.; Repetto, D.; Sessolo, M.; Barea, E. M.; Bisquert, J.; Garcia Belmonte, G.; Prochazka, J.; Kavan, L. Adv. Funct. Mater. 2008 , 18 , 145 150. (111) Xiao, X.; Zimmerman, J. D.; Lassiter, B. E.; Bergemann, K. J.; Forrest, S. R. Appl. Phy s. Lett. 2013 , 102 , 073302 . (112) Yip, H. L.; Hau, S. K.; Baek, N. S.; Ma, H.; Jen, A. K. Y. Adv. Mater. 2008 , 12 , 2376 2382.
129 (113) Hau, S. K.; Yip, H. L.; Ma, H.; Jen, A. K. Y. Appl. Phys. Lett. 2008 , 93 , 233304. (114) Langmuir 2001 , 17 , 4014 4020. (115) Tosatti, S.; Michel, R.; Textor, M.; Spencer, N. D. Langmuir 2002 , 18 , 3537 3548. (116) Hanson, E. L.; Schwartz, J.; Nickel, B.; Koch, N.; Danisman, M. F. J. Am. Chem. Soc. 2003 , 125 , 16074 16080. (117) Zhang, H.; Zhang, L.; Wang, J. J. Am. Chem. Soc . 2009 , 131 , 13206 13207. (118) Zhang, H.; Xu, D.; Wang, J. Mater. Chem. Phys. 2012 , 133 , 655 660. (119) Vempati, S.; Ertas, Y.; Uyar, T. J. Phys. Chem. C 2013 , 117 , 21609 21618. (120) Sharma, M.; Kumar, S.; Pandey, O. P. Dig. J. Nanomater. Bios. 2008 , 3 , 189 197. (121) Maturana, H . A.; Peric, I. M.; Pooley, S. A.; Rivas, B. L. Polym. Bull. 2000 , 45 , 425 429. (122) Lahiri, S.; Sarkar, S. Appl. Radiat. Isotopes 2007 , 65 , 387 391. (123) Xiao, L.; Chen, Z.; Qu, B.; Luo, J.; Kong, S.; Gong, Q.; Kido, J. Adv. Mater. 2011 , 23 , 926 952. (124) Chopra, N.; Le e, J.; Zheng, Y.; Eom, S. H.; Xue, J.; So, F. Appl. Phys. Lett. 2008 , 93 , 143307. (125) Kim, M.; Lee, J. Y. Org. Electron. 2012 , 13 , 1245 1249. (126) K. W. Appl. Phys. Lett. 2008 , 92 , 113505. (127) Chiang, C. L.; Tseng, S. M.; Chen, C. T.; Hsu, C. P.; Shu, C. F. Adv. Funct. Mater. 2008 , 18 , 248 257. (128) Chen, S.; Manders, J. R.; Tsang, S. W.; So, F. J. Mater. Chem. 2012 , 22 , 24202 24212. (129) Oyama, H. T.; Wightman, J. P. Surf. Interface Anal. 1998 , 26 , 39 55. (130) Darling, R. B.; IwanÃ¤ga, S. Sadhana 2009 , 34 , 531 542. (131) Xiao, Z.; Bi, C.; Shao, Y.; Dong, Q.; Wang, Q.; Yuan, Y.; Wang, C.; Gao, Y.; Huang, J . Energy Environ. Sci. 2014 , 7 , 2619 2623. (132) Schmidbauer, S. ; Hohenleutner, A. ; B. Beilstein J. Org. Chem . 2013 , 9 , 2088 2096.
130 (133) Kondakov, D. Y. ; Lenhart, W. C. ; Nichols, W. F. J. Appl. Phys. 2007 , 101 , 024512. (134) Popovic, Z. D. ; Aziz, H. ; Hu, N. X. ; Ioannidis, A. ; dos Anjo, P. N. M. J. Appl. Phys. 2001 , 89 , 4673. (135) Schmidt, T. D.; Setz, D. S.; FlÃ¤mmich, M.; Scholz, B. J.; Jaeger, A.; Diez, C.; Michaelis, D.; Danz, N.; BrÃ¼tting, W. Appl. Phys. Lett. 2012 , 101 , 103301. (136) Giebink,N. C. ; ; Weaver,M. S. ; Brown, J. J. ; Forrest,S. R. J. Appl. Phys. 2009 , 105 , 124514. (137) Nowy, S. ; Ren, W. ; Elschner, A. ; ; BrÃ¼tting, W. J. Appl. Phys. 2010 , 107 , 054501. (138) Kondakov, D. Y.; J. Appl. Phys. 2008 , 104 , 084520 . (139) Xiang, C.; Chopra, N.; Wang, J.; Brown, C.; Ho, S.; Mathai, M.; So , F. Org. Electron. 2014 , 15 , 1702 1706. (140) Chopra, N. SID Symp. Dig. Tech. 2012 , 43 , 438 440. (141) Kim, J. H.; Chen, Y.; Liu, R.; So, F. Org. Electron. 2014 , 15 , 2381 2386. (142) Yamamoto, H. ; Brooks, J. ; Weaver, M. S. ; Brown, J. J. ; Murakami, T. Murata, H. Appl. Phys. Lett. 2011 , 99 , 033301. (143) Kondakova, M. E.; Pawlik, T. D.; Young, R. H.; Giesen, D. J.; Kondakov, D. Y. J. Appl. Phys. 2008 , 104 , 094501. (144) Reineke, S.; Walzer, K.; Leo, K. Phys. Rev. B , 2007 , 75 , 125328 . (145) Song, D.; Zhao, S.; Luo, Y.; Aziz, H. Appl. Phys. Lett. 2010 , 97 , 243304. (146) van Eersel, H. ; Bobbert, P. A. ; Janssen, R. A. J. ; Coehoorn, R. Appl. Phys. Lett. 2014 , 105 , 143303. (147) Zhang, Y. ; Lee, J. ; Forrest, S. R. Nat. Commun. 2014 , 5 , 5008. (148) Yook, K. S.; Lee, J. Y. Adv. Mater. 2014 , 26 , 4218 4233. (149) Griesbeck, A. G.; Cho, M. Org. Lett . 2007 , 9 , 611 613.
131 BIOGRAPHICAL SKETCH The author got a bachelor degree from Department of Materials Science and Engineering, Tsinghua University in June 2010 Department of Ma terials Science and Engineering, University in Aug. 2011. His interests lie in solution process able organic and hybrid optoelectronic devices with emphasis on the novel device architecture and fabrication approach. He received his Ph.D . degree from the Un iversity of Florida in the spring of 2015 .